Output properties of diode-pumped passively Q-switched 1.06 μm Nd:GdVO₄ laser using a [100]-cut Cr⁴⁺:YAG crystal

Abstract

A passively Q-switched 1.06 μm Nd:GdVO₄ laser with a [100]-cut Cr⁴⁺:YAG saturable absorber was demonstrated. The output characteristics were investigated when the anisotropic transmission of Cr⁴⁺:YAG crystal and the incident pump power level were considered. The experimental results showed that it was feasible to generate laser with narrower pulse width (τ _p ), higher pulse energy and peak power when the polarization direction of laser was parallel to the [001], [010], [$00\overline{1}$], and [$0\overline{1}0$] orientations of the Cr⁴⁺:YAG crystal. The different changes of τ _p as a function of incident pump power was observed due to the anisotropy of transmission of Cr⁴⁺:YAG and the different gain levels (pump power levels). If the Cr⁴⁺:YAG was fully bleached as a result of high cavity gain or due to the laser polarization direction was parallel to the [001], [010], [$00\overline{1}$], and [$0\overline{1}0$] orientations, τ _p was constant, otherwise τ _p decreased when the gain increased.

1 Introduction

Q-switched lasers are used in many applications, such as laser lidar, remote sensing, micro-machining, and microsurgery [1, 2]. Compared with active Q-switching, the passive techniques have advantages such as lower cost, compactness, simplicity in set-up, and operation since they do not require external control. Cr⁴⁺:YAG crystal as a saturable absorber has advantages of improved thermo-mechanical properties, large absorption cross section, low saturable intensity, and high damage threshold [3, 4].

The output performance, such as output energy, E _p , pulse duration, τ _p , pulse peak power, P _p , and pulse repetition rate, f, are important for passively Q-switched lasers. Many theoretical models and experimental results about passively Q-switched lasers have appeared in the literature [5–12]. But some of the theoretical opinions and experimental observations differed in these publications. For example, Degnan [5] reported that τ _p was not affected by the pump power level, but was determined by other parameters, such as the cavity length. This was confirmed by experimental results reported in [6–8]. However, Lu et al. [9] reported that τ _p decreases when the incident pump power increases, which was supported by experimental investigations [10–12]. In our characterization studies of a Cr⁴⁺:YAG passively Q-switched Nd:GdVO₄ laser, both these two different experimental phenomenon were observed for the first time in the same experimental configuration, which we attributed them to the anisotropy of the Cr⁴⁺:YAG transmission and the different laser cavity gain. The anisotropic transmission of Cr⁴⁺:YAG crystal was previously reported in [13, 14], but to the best of our knowledge, there was no researcher to report the above two changes of τ _p simultaneously, not to mention giving an explanation.

In this paper, we demonstrated a passively Q-switched 1.06 μm Nd:GdVO₄ laser with a [100]-cut Cr⁴⁺:YAG crystal by 808 nm laser diode pumping. The output characteristics were investigated when the anisotropic transmission of Cr⁴⁺:YAG crystal and the different gain levels were considered. The experimental results showed that we could obtain a pulsed laser output with narrower τ _p , higher E _p and P _p when the polarization direction of laser was parallel to the [001], [010], [$00\overline{1}$], and [$0\overline{1}0$] orientations of the Cr⁴⁺:YAG crystal, respectively. The different changes of τ _p as a function of incident pump power were observed due to the anisotropic transmission of Cr⁴⁺:YAG crystal and the different cavity gain levels.

2 Experimental setup

The experimental configurations of a [100]-cut Cr⁴⁺:YAG passively Q-switched Nd:GdVO₄ laser using 808 nm laser-diode end-pumping is shown in Fig. 1.

Fig. 1

Experimental setup of a passively Q-switched Nd:GdVO₄ laser and the schematic of crystallographic axis of the Cr⁴⁺:YAG, polarized laser, and a rotation angle β

The 808 nm pump source was a fiber coupled laser diode. L₁ and L₂ were a set of collimating and focusing lenses. The a-cut Nd:GdVO₄ crystal with a 0.3 at.% Nd³⁺ concentration had dimensions of 3×3×8 mm³. The crystal was wrapped with indium foil and placed into water-cooled copper heat sink. The input mirror M₁ was a flat mirror with an antireflection coating at 808 nm and a high reflectivity at 1063 nm. The output coupler M₂ had a transmission of 30 % at 1063 nm. The [100]-cut Cr⁴⁺:YAG crystal with initial transmission T ₀ of 90 % acted as a saturable absorber. The cavity length was about 160 mm.

3 Experimental results and discussions

The dependence of energy transmission T _i of the Cr⁴⁺:YAG crystal with the angle between the crystallographic axis and the polarization direction of laser was investigated firstly. The crystal was excited with a linearly polarized, AO Q-switched 1063 nm Nd:GdVO₄ laser propagating along its growth direction. The schematic of crystallographic axis of Cr⁴⁺:YAG, polarized laser and rotation angle β is same as shown in Fig. 1. The laser was focused by a lens to increase the power density I _L . The crystal was fixed on a precise rotatable mount in order to measure the rotation dependent T _i . The experimental results are shown in Fig. 2.

Fig. 2

T _i of the [100]-cut Cr⁴⁺:YAG crystal as a function of rotation angle

From Fig. 2, we can see that at a power density of I _L =4×10⁵ W/cm², there is an obvious anisotropic energy transmission. T _i varied with the angle β, and the almost same T _i peaks appeared every 90^∘. When β was 0^∘, 90^∘, 180^∘, and 270^∘, the polarization direction of laser was parallel to the [001], [010], [$00\overline{1}$], and [$0\overline{1}0$] orientations of Cr⁴⁺:YAG, respectively, and T _i had the maximum value. When the polarization direction of laser was parallel to the [011], [$01\overline{1}$], [$0\overline{1}\overline{1}$], [$0\overline{1}1$], orientations (β=45^∘, 135^∘, 225^∘, and 275^∘, respectively), T _i was minimum.

The output characteristics of passively Q-switched Nd:GdVO₄ laser with the [100]-cut Cr⁴⁺:YAG crystal were investigated. The experimental configuration is shown in Fig. 1. The incident pump power was 7.5 W and the Cr⁴⁺:YAG crystal was rotated in a full 360^∘. The measured results are shown in Figs. 3–7.

Fig. 3

Relation between the average output power and the rotation angle β

From Fig. 3, we can see that the maximum average output power was obtained when β was 0^∘, 90^∘, 180^∘, and 270^∘, respectively. This is because of when the polarization direction of laser was parallel to the [001], [010], [$00\overline{1}$], and [$0\overline{1}0$] orientations, the T _i of Cr⁴⁺:YAG crystal had a maximum value. The loss was smaller and the laser was easier to oscillate compared with the directions of [011], [$01\overline{1}$], [$0\overline{1}\overline{1}$], and [$0\overline{1}1$].

The modulation depth Q _o of Cr⁴⁺:YAG crystal is approximately directly proportional to the difference of T _i and T ₀. Q _o peaks emerged when β was 0^∘, 90^∘, 180^∘, and 270^∘, respectively. Therefore, at these β values, the repetition rate and the pulse width were at a minimum and the pulse energy and the pulse peak power were maximum, as shown in Figs. 4, 5, 6 and 7.

Fig. 4

Relation between the repetition rate and the rotation angle β

Fig. 5

Relation between the pulse width and the rotation angle β

Fig. 6

Relation between the pulse energy and the rotation angle β

Fig. 7

Relation between the pulse peak power and the rotation angle β

There are four sinusoidal modulations of T _i in a full 360^∘. We studied the pulse width as a function of incident pump power at different rotation degree within 45^∘ to illustrate the law of the full 360^∘. The measured results are depicted in Fig. 8.

Fig. 8

Pulse width as a function of incident pump power at different rotation degree

From Fig. 8, we can see that at the same incident pump power, the narrowest τ _p was achieved when β was 0^∘ and the largest could be obtained when β was 45^∘. Due to the anisotropic transmission, at β of 0^∘ and in the investigated pump power range (from 7.5 W to 13.7 W), the Cr⁴⁺:YAG crystal was fully bleached, and the pulse width was constant and ∼76 ns. At other angles (β=10^∘, 22.5^∘, 35^∘, and 45^∘, respectively), τ _p was a decreasing function of the incident pump power, and it would decrease to a constant value of 76 ns (β=0^∘) when the incident pump power increased. Because the gain of the laser oscillator is a function of the incident pump power level, as shown in Fig. 8, the gain impacts the pulse duration. If the Cr⁴⁺:YAG was fully bleached as the result of high gain or due to the laser polarization direction was parallel to the [001], [010], [$00\overline{1}$], and [$0\overline{1}0$] orientations, τ _p was constant, otherwise τ _p decreased when the gain increased.

4 Conclusions

In conclusion, we have demonstrated a passively Q-switched 1.06 μm Nd:GdVO₄ laser with a [100]-cut Cr⁴⁺:YAG saturable absorber by 808 nm laser diode pumping. The output characteristics were investigated when the anisotropic transmission of Cr⁴⁺:YAG crystal and the different gain levels were considered. The experimental results showed that a pulsed laser output with narrower τ _p , higher E _p and P _p can be achieved when the laser polarization direction was parallel to the [001], [010], [$00\overline{1}$], and [$0\overline{1}0$] orientations of Cr⁴⁺:YAG, respectively. Two different effects of constant τ _p and varying τ _p were observed due to the anisotropy of transmission of Cr⁴⁺:YAG crystal and the gain level. If the Cr⁴⁺:YAG was fully bleached as the result of high gain or due to the laser polarization direction was parallel to the [001], [010], [$00\overline{1}$], and [$0\overline{1}0$] orientations of Cr⁴⁺:YAG, which was more easily to be complete bleached, τ _p was constant, otherwise τ _p decreased when the gain increased.