Open Access
Add to BibSonomy
Abstract
Here we demonstrate an efficient, long-distributed-cavity laser which uses a cat-eye retroreflector configuration to facilitate cavity alignment. The cavity parameters were optimized to meet the small stability region of the cavity, given the long working distance. We also found that intracavity spherical aberration seriously impacts laser efficiency, and an aspheric lens is used to correct the aberration when long working distance and receivers with compact dimensions are desired. The end-pumped Nd:YVO4 laser delivered 5.91 W continuous-wave output power at a long working distance of 5 m, under 16.6 W incident diode pump. Significantly, the fluctuation in output power over the whole working distance range of 1–5 m was less than 10%, this being achieved without any other cavity parameters being adjusted.
© 2021 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
Lasers with a distributed cavity arrangement typically feature a total reflector and an output coupler which are spatially separated by more than is typical (hundreds of millimeters or shorter) in a “standard” laser resonator. Such lasers have applications in laser communication, laser charging, sensing and distance measurement [1–4]. For applications like laser charging and laser communication, the laser output is delivered directly onto a photovoltaic (PV) cell and/or the detector, which is closely-coupled with the laser output coupler [1,2]. This removes the necessity of aiming and tracking which is typical when using a conventional laser source. If an item which is opaque at the laser wavelength is positioned within the light-path, it will prevent the distributed cavity laser from oscillating. This can be used as a means of laser safety [1], and as a means of intracavity sensing of media such as gases (which have absorption peaks at the laser wavelengths) [3].
Usually, a free space laser with a cavity length of several meters is challenging to align, and typically, retroreflectors (which can reflect an incident beam back 180°) are used to aid cavity alignment. The corner cube is one of the most commonly used retroreflectors. Laser oscillation has been demonstrated in Xenon and Nd:YAG lasers employing corner cubes, with cavities as long as 30 km and 6.3 km, respectively [5,6]. When using a long distributed cavity, the laser beams also need to be well collimated to overcome the divergence of the Gaussian beam given the long working distance. The main drawback of the corner cube retroreflector is that its edges and center do not provide the highest reflection, and this can result in enormous loss and development of undesired transverse modes [7,8]. Also, the backward-propagating beam usually suffers a transverse shift upon multiple reflections, and loses overlap with the incident beam. This somewhat limits its application to incorporation with gain media that have large cross-sectional area [9,10]. Therefore, lasers which incorporate corner cube retroreflectors are typically high-gain and pulsed, generating high-energy output. For the aforementioned applications, stable continuous-wave (CW) lasers are preferred, and hence, corner cubes are not the ideal retroreflector.
Another kind of retroreflector which can be used in distributed lasers is a cat-eye optical arrangement. In this arrangement, a mirror is placed at the focal plane of a focusing lens; this results in light which passes through the pupil from any angle of incidence within the field of view of the lens, being reflected back along the incident path, thereby facilitating cavity alignment. In 2005, Xu et. al. demonstrated a CW He-Ne laser in which the highly-reflective mirror was replaced by a cat-eye optic. The laser, with a cavity length of over 1.1 m, exhibited outstanding resistance to misalignment [11]. Zeng et. al. used a cat-eye optic and a corner cube to form the cavity of a pulsed Nd:YAG laser. Laser oscillation was observed with a cavity length of over 5 m. The robustness of the design was demonstrated by the authors using photo paper and a CD ROM (both of which provided very limited reflection) as the reflector in the cat-eye [12]. External-cavity diode lasers and copper vapor lasers using cat-eye optics with good resistance to misalignment have also been reported [13,14]. Cat-eye retroreflectors have therefore proven very effective at enhancing misalignment tolerance of laser cavities, and thus, could facilitate the alignment of long, distributed-cavity lasers. To the best of our knowledge, efficient CW operation of a distributed-cavity laser with long working distance has not been demonstrated.
In this work, we present a CW long-distributed-cavity Nd:YVO4 laser using two cat-eye retroreflectors within the cavity. The cavity stability region of the fundamental transverse mode was optimized for efficient laser operation over a large cavity length (working distance) range. We also found that spherical aberration (SA) plays an important role in laser efficiency, especially when a long laser working distance requires a large collimated beam size. After optimizing the cavity and correcting for spherical aberration, over 5 W laser output power at 1064 nm was obtained across a working distance range of 1 m to 5 m, under an incident 808 nm diode pump power of 16.6 W. The laser output power is also quite insensitive to changes in cavity working distance, suggesting this may be a promising source for laser charging applications.
2. Experiment arrangement
The design of the distributed cavity laser is shown in Fig. 1. The pump source is a fiber-coupled laser diode (LD) emitting at a wavelength of 808 nm. The pump light is delivered via an optical fiber with a core diameter of 400 µm and a numerical aperture of 0.22. This is collimated and refocused into the laser crystal (LC) with a beam radius of 500 µm by a pair of focusing lenses. The LC is a 4 mm long, 0.5 at.% doped, a-cut Nd:YVO4 crystal, which absorbs ∼85% of the incident unpolarized pump light. A focusing lens F1 with a focal length of 51.8 mm at the 1064 nm laser wavelength and a flat highly reflective (HR) mirror M1 (R>99.9% at 1064 nm) formed a cat-eye retroreflector. Another focusing lens F4 combined with a curved T=20% output coupler M2 formed the receiver cat-eye. Multiple F4 lenses with different focal lengths were investigated in the experiment. Two lenses F2 and F3 with focal lengths of 51.8 mm and 100 mm, respectively, made an intracavity telescope to expand and collimate the laser beam for efficient laser operation over a long working distance. A flat mirror MF, which is coated HR at 1064 nm and anti-reflective (AR) at 808 nm at 45° angle of incidence, was used to fold the cavity so that the beam profile could be monitored by a CCD camera (Ophir SP907) behind M1. The distance between the camera and M1 was equal to that between M1 and F1. Since the laser beam waist was located at the flat total reflector M1, the beam size at camera was equal to that at lens L1, and was similar to that at the laser crystal (designed to be 450–500 µm to match the pump size). All the intracavity components, including the laser crystal and the lenses, are coated AR at the 1064 nm laser wavelength (R<0.2%). The Nd:YVO4 crystal was also coated AR at 808 nm (R<3%) on both facets. This dual-cat-eye cavity arrangement was able to tolerate over 2° misalignment of the receiver in the experiment (with the laser output power maintaining more than 50% of its maximum value when the whole receiver tilted within this range), making alignment of the long-distributed-cavity relatively simple.
Fig. 1. Schematic of the distributed cavity laser. The definitions of the distances between each component marked in the figure are as follows. L1: M1-F1; L2: F1-MF-LC; L3: LC-F2; L4: F2-F3; L5 (working distance): F3-F4; L6: F4-M2.
For conciseness, we use L1-L6 to describe the distance between each component forming the distributed-cavity laser, as shown in the caption of Fig. 1. In this distributed-cavity laser (as designed for laser charging applications), the transmitter comprises optics M1 to F3, which incorporate the pump source and the coupling optics. Meanwhile, the receiver cat-eye retroreflector comprises optics F4 and M2, and the “working distance” refers to distance between the transmitter and the receiver (i. e., the distance between F3 and F4, namely, L5). In designing the laser, one of the key objectives was to ensure that the laser could operate efficiently (maintaining output power and good transverse mode profile) over a large working distance range of several meters. This required the fundamental mode size at the Nd:YVO4 crystal to be well matched with the pump beam size over the whole working distance range.
The thermal focal length Ft in the 0.5-at.%-doped Nd:YVO4 crystal was estimated to be ∼300 mm using the equation in Ref. [15], this with the laser operating with the maximum pump power of 16.6 W, and with a beam radius of ∼500 µm. The distances between each component of the transmitter L1, L2, L3 and L4 were 48 mm, 51 mm, 51 mm and 148 mm, respectively. Figure 2 shows the calculated stability region of L6 in the receiver cat-eye versus working distance L5, when using a f=51.8 mm lens as F4. It can be seen that the stable region of L6 narrowed significantly from several millimeters to ∼1 mm when the working distance L5 increased from 0.5 m to 5 m. Hence, the distance L6 must be carefully optimized based on the working distance (L5) required. Figure 3 plots the TEM00 mode radii in the laser cavity with a 5 m long working distance. The beam has very small beam waists of a few tens of microns at the flat total reflector M1, between F2 and F3, and near the output coupler M2. Meanwhile, the beam is collimated between the lenses F1 and F2, and between F3 and F4. The beam radius at F3 is ∼1 mm, with corresponding Rayleigh length of 3 m, which makes the long working distance possible.
Fig. 2. Stability region (shaded) of L6 with different working distance L5. L1 = 48 mm, L2=L3 = 51 mm, L4 = 148 mm, Ft=300 mm. The thicknesses of the lenses are not considered.
Fig. 3. The TEM00 mode radius in the distributed laser cavity (L1 = 48 mm, L2=L3 = 51 mm, L4 = 148 mm, Ft=300 mm, L6 = 52.2 mm. The thicknesses of the lenses are not considered). Inset: zoomed-in portion of the plot, in the region of the transmitter.
3. Results and discussion
3.1 Influence of L6 on working distance range
The experiment was carried out with the arrangement detailed in the above section. We first used a receiver cat eye comprising a f=51.8 mm lens and an output coupler with a ROC of 51.8 mm, equal to the focal length of F4. Figure 4 shows the power transfer characteristic of the distributed cavity laser. The 1064 nm laser output power under the maximum incident diode pump power of 16.6 W were 6.39 W, 6.22 W and 5.2 W, at a working distance of 1 m, 2 m and 5 m respectively. The optical efficiency of the distributed cavity laser with respect to the incident pump power decreased slightly from 38.5% to 31.3% as the working distance increased from 1 m to 5 m. Due to the retroreflection of the cat-eye optics, the alignment of the receiver was a simple process.
Fig. 4. Laser output power as a function of diode pump power, with different working distance L5 of 1 m, 2 m and 5 m. The lines are a guide to the eye.
It should be mentioned that the data in Fig. 4 was obtained with the distance L6 optimized for each working distance. When considering practical applications, a distributed cavity laser should be able to operate efficiently over the working distance range required, with the receiver maintaining fixed parameters, rather than having to adjust L6 for different working distances L5. When keeping the distance L6 unchanged for different L5, we found that the laser characteristics would change dramatically. As shown in Fig. 5, when L6 was optimized for a relatively short working distance of 2 m, the output power under a fixed incident pump power of 16.6 W would drop significantly when increasing the working distance L5 (black empty square). In contrast, if L6 was optimized for a longer working distance of 5 m, the laser output power would not change a lot when the working distance became shorter (solid red circle). The optimized distance for L6, producing maximum output power at 5 m working distance was ∼0.5 mm smaller than that optimized at 2 m. This can be understood by examining Fig. 2 which shows the stable region of L6 – it is wide for short values of L5 but becomes much narrower when L5 gets longer. Using the CCD camera, we observed a beam radius of ∼450 µm in the laser crystal when L6 was optimized at L5 = 2 m, which matched the pump beam size well. As the working distance increased, the beam size expanded rapidly. However, when L6 was optimized for L5 = 5 m, the cavity could maintain stability as the working distance decreased, with the beam size at the laser crystal not changing significantly. The laser output power remained relatively stable during this reduction in working distance L5.
Fig. 5. Laser output power as a function of working distance L5 under 16.6 W incident diode pump power, with L6 optimized at L5 = 2 m, L5 = 5 m, and at each point, respectively. The lines are a guide to the eye.
3.2 Influence of and compensation of the spherical aberration
Receivers with small overall size are usually preferred in practical applications of distributed cavity lasers. Since the receiver size is mainly determined by the distance L6 which in itself is a function of focal length F4, we examined the efficacy of lenses F4 with shorter focal lengths of f=33.9 mm and f=25 mm. The focal length, substrate material and shape of the lenses used are given in Table 1. Figure 6 plots the laser output power as a function of working distance with different lenses. The blue diamond represents the output power with an f=51.8 mm lens (lens #1) and L6 optimized at each working distance (which is the same configuration used to produce the blue diamond curve in Fig. 5). When using the f=33.9 mm spherical lens (lens #2), the laser output power was lower than that achieved with the f=51.8 mm lens. Even lower output power was recorded with the f=25 mm spherical lens (lens #3). It can be seen that the decrease on output power became much more serious with longer working distance. Under the fixed incident pump power of 16.6 W, the laser output power with when using F4 spherical lenses with focal lengths of 51.8 mm, 33.9 mm and 25 mm was 6.39 W, 6.34 W and 6.22 W respectively, at a working distance of 1 m. These values decreased to 5.2 W, 2.58 W and 1.3 W respectively when the working distance was increased to 5 m.
Fig. 6. Laser output power as a function of working distance L5 using spherical F4 lenses with different focal lengths, and an aspheric F4 lens with L6 optimized at each working distance. Also shown is the output power when using an aspheric F4 lens with fixed L6, optimized for a working distance of 5 m and for 16.6 W incident pump power. The lines are a guide to the eye.
Table 1. Parameters of the lenses used as F4
It is well known that spherical mirrors and lenses can introduce spherical aberration in an optical system. The influence of SA is usually neglected in end-pumped solid-state lasers due to the associated small beam size. However, in the distributed-cavity laser investigated here, the beam expands at the receiver lens F4, given long working distances, thus resulting in large amounts of SA. The rays at the edge of the wide beam undergo stronger focusing than that of the central rays passing through the lens. Therefore, the actual focal point of the edge rays deviate from that of the central rays, or nominal focal point of the lens (it is this deviation in focal point that we use in this work to quantify the effect of SA). In this case, when optimizing the location of the output coupler M2 (i. e., length L6) for optimal feedback of the central rays, the rays at the edge of the beam would effectively suffer some loss. In other words, due to the effect of SA, the cat-eye optics do not provide ideal retroflection of the entire beam. Typically spherical lenses with short focal length suffer strong SA. We believe that this is why the laser output power decreased when using the 33.9 mm and 25 mm focal length lenses instead of the 51.8 mm focal length lens. Another interesting observation related to SA is that the output beam profile may become donut-like when the M2 is deviated from its optimal location - since the higher-order Laguerre-Gaussian (LG) modes have beam radii large than that of the TEM00 mode, their actually focal points are deviated as a result. Comprehensive study on LG modes generation based on this has been presented in another work [16].
We calculated the TEM00 mode radius at F4 as a function of working distances (using ABCD matrix), as well as the resultant SA of each lens (using Zemax software); this is plotted in Fig. 7. The SA is plotted in terms of the deviation between the actual focal points of the edge rays and the nominal focal points of the lenses. We can see that the fundamental mode beam radius at F4 was ∼930 µm at 1 m working distance, and this expanded to ∼2 mm when L5 increased to 5 m. As a result, the SA of the f=51.8 mm spherical lens increased from 0.19 mm to 0.34 mm. For the f=33.9 mm and f=25 mm spherical lenses, the SA at 5 m working distance reached 0.62 mm and 0.70 mm, respectively. We can see from Fig. 2 that the stable region of L6 was less than 1 mm at 5 m working distance. For a Gaussian beam, there is still 13.5% of the total energy out of the beam radius. Therefore, such a large SA would prevent a large part of light from being retroreflected, and seriously impact the laser efficiency.
Fig. 7. Calculated fundamental beam radius at F4 as a function of working distance L5, and the corresponding spherical aberration (quantified as the deviation in focal position of edge rays to central rays which pass through lens F4) when using each lens.
To compensate the effect of SA, we used an f=25 mm aspheric lens (lens #4, Edmund 89–439) as F4. The aspheric lens has a diameter of 25 mm and a clear aperture (CA) of 23 mm, which are sufficient for the incident beam. The results achieved when using this lens, and with L6 optimized at each working distance were also plotted in Fig. 6 (the red un-filled circle). After correcting the SA in receiver cat-eye, the laser output power at longer working distance improved significantly. The maximum output power at L5 = 5 m reached 5.91 W, over 4 times higher than that when using the spherical lens with the same focal length (#3), and was even higher than the case of the f=51.8 mm spherical lens (#1) which exhibits moderate SA. Correcting the aberration makes the distributed cavity laser capable of operating efficiently at long working distance, and with a compact receiver size. We also recorded the laser output power as a function of working distance L5 with a fixed L6, optimized at L5 = 5 m; this data is plotted as the solid gray circles in Fig. 6. The maximum and minimum output power of 6.22 W and 5.73 W were obtained at the working distances of 2.5 m and 4 m, respectively, and the power fluctuation over the whole working distance range of 1–5 m was less than 10%. For the transmitter cat-eye, since the beam radius at the lens F1 was only ∼450 µm, the calculated SA of F1 was as small as ∼0.02 mm, which is negligible compared with that at the receiver end, and correction is therefore unnecessary. With the aspheric lens, the SA is no longer a problem and the laser should be capable of working efficiently at even longer working distance. However, the collimated beam size at the transmitter should be selected considering the Rayleigh length and the beam size at the receiver, given the limited lens aperture. In the experiment we also tried another aspheric lens Thorlabs C260TMD-C, which also corrected the SA well. However, the small CA of 5.0 mm limited the output power significantly at long working distance. For further power scaling, the maximum pump power of 16.6 W used in this proof of principle experiment is still far below the damage threshold of the laser crystal [17]. Therefore, output power at ten-watt level or higher can be obtained by simply increasing the pump power and pump beam size.
It is also worth mentioning that the working distance is not the only important issue for practical applications given in the introduction. The misalignment tolerance mentioned in the section 2, i. e., the field of view (FoV), determines if the laser could work “alignment-free” over a large range. This is limited by another type of aberration, the field curvature. The results of optimizing the FoV to ±20° by compensating the field curvature will be presented in another work.
4. Conclusion
In conclusion, we have demonstrated an efficient, long-distributed-cavity laser using cat-eye retroreflectors. We found that the cavity stability region becomes very narrow at long working distances, and this places strict requirements on the distances between the elements, particularly within the receiver. It was found that SA significantly impacts the output power of the system, given the narrow cavity stability range, and large beam sizes involved. By correcting the SA, our end-pumped, distributed cavity, Nd:YVO4 laser could deliver 5.91 W CW output power, under 16.6 W incident pump, with a working distance of 5 m. The output power could be maintained above 5.7 W for variable working distances in the range of 1–5 m without other cavity parameters being adjusted. We believe that the long working distances and robustness makes this distributed-cavity laser design a potential source for practical applications such as distributed laser charging.
Funding
National Natural Science Foundation of China (61975146).
Acknowledgement
Quan Sheng thanks Dr. Mingqiang Wang and Dr. Zefeng Lu (Huawei Technologies Co., Ltd.) for helpful discussions.
Disclosures
The authors declare no conflicts of interest.
Data availability
Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.
References
1. Q. Zhang, W. Fang, Q. Liu, J. Wu, P. Xia, and L. Yang, “Distributed Laser Charging: A Wireless Power Transfer Approach,” IEEE Internet Things J. 5(5), 3853–3864 (2018). [CrossRef]
2. M. Xiong, Q. Liu, G. Wang, G. Giannakis, and C. Huang, “Resonant beam communications: principles and designs,” IEEE Commun Mag 57(10), 34–39 (2019). [CrossRef]
3. E. Mehdizadeh, J. Lunine, and G. Atkinson, “Intracavity laser spectroscopy with an ion-doped, solid-state Tm3+:YAG laser,” J Quant. Spectrosc. Ra. 68(4), 453–465 (2001). [CrossRef]
4. A. S. Medovikov, “Some design principles of interference range finders,” Meas Tech 28(11), 919–922 (1985). [CrossRef]
5. G. Linford, E. Peressini, W. Sooy, and M. Spaeth, “Very long lasers,” Appl. Opt. 13(2), 379–390 (1974). [CrossRef]
6. G. Linford and L. Hill, “Nd:YAG long lasers,” Appl. Opt. 13(6), 1387–1394 (1974). [CrossRef]
7. G. Zhou and A. Alfrey, and. Casperson, “Modes of a laser resonator with a retroreflecting corner cube mirror,” Appl. Opt. 21(9), 1670–1674 (1982). [CrossRef]
8. M. Shen, S. Wang, L. Hu, and D. Zhao, “Mode properties produced by a corner-cube cavity,” Appl. Opt. 43(20), 4091–4094 (2004). [CrossRef]
9. Y. Wang, T. Dai, X. Liu, Y. Ju, and B. Yao, “Dual-wavelength injection-seeded Q-switched Ho:YLF laser for CO2 differential absorption lidar application,” Opt. Lett. 44(24), 6049–6052 (2019). [CrossRef]
10. Y. Tan and S. Zhang, “Alignment-free He–Ne laser with folded cavity,” Opt Laser Technol 46(8), 578–581 (2008). [CrossRef]
11. Z. Xu, S. Zhang, Y. Li, and W. Du, “Adjustment-free cat's eye cavity He-Ne laser and its outstanding stability,” Opt. Express 13(14), 5565–5573 (2005). [CrossRef]
12. Q. Zeng, Y. Wan, and K. Qin, “Laser resonator of novel configuration applicable to efficient electro-optical countermeasure,” in Chinese) Infrared Laser Eng. 46(8), 806002 (2017). [CrossRef]
13. B. Singh, S. Daultabad, V. Subramaniam, and A. Chakraborty, “Performance of a 80W copper vapor laser with “alignment free” unstable CAT-EYE resonator and other configurations using intra-cavity apertures,” Optics Commun. 281(24), 6080–6090 (2008). [CrossRef]
14. B. Fermigier, G. Lucas-Leclin, J. Dupont, F. Plumelle, and M. Houssin, “Self-aligned external-cavity semiconductor lasers for high resolution spectroscopy,” Opt. Commun. 153(1-3), 73–77 (1998). [CrossRef]
15. M. Innocenzi, H. Yura, C. Fincher, and R. Fields, “Thermal modeling of continuous-wave end-pumped solid-state lasers,” Appl. Phys. Lett. 56(19), 1831–1833 (1990). [CrossRef]
16. M. Wang, Y. Ma, Q. Sheng, X. He, J. Liu, W. Shi, J. Yao, and T. Omatsu, “Laguerre-Gaussian beam generation via enhanced intracavity spherical aberration,” Opt. Express 29(17), 27783–27790 (2021). [CrossRef]
17. Y.-F. Chen, “Design criteria for concentration optimization in scaling diode end-pumped lasers to high powers: influence of thermal fracture,” IEEE J. Quantum Electron. 35(2), 234–239 (1999). [CrossRef]
