Development of Single-Longitudinal-Mode Selection Technology for Solid-State Lasers

Lasers with narrow linewidths and single frequencies are widely used in ﬁelds such as radar detection, nonlinear optics, and precision measurements. The demand for such lasers has promoted the rapid development of single-longitudinal-mode (SLM) selection technology. Here, we highlight the working principles of current mainstream SLM selection technologies and the recent advances in the ﬁeld. We compare the characteristics of diﬀerent SLM selection methods and list the challenges faced by these technologies.


Introduction
With only one fundamental frequency component, singlelongitudinal-mode (SLM) lasers have the advantages of narrow linewidths and low phase noise and are therefore ideal for applications where a high-coherence light source is required, including Doppler wind lidar, atmospheric composition measurement, coherent optical communication, gravitational wave detection, sodium guide star, and nonlinear optics [1][2][3][4][5]. In general, if the linewidth of an SLM laser is narrower than the resolution of a spectrum analyzer or the nonlinear gain bandwidth of a specific medium, it is assumed to be a single-frequency laser. e generation of SLM lasers is usually based on a combination of the linewidth of the gain medium and the losses of the cavity. Alternatively, SLM selection methods can be applied to solid state lasers to generate SLM lasers [6,7]. Owing to their strong power (average and peak) scaling ability and wide range of wavelength coverage, solid-state lasers play an important role in space exploration, defense, and manufacturing [8][9][10]. In the past few decades, there has been a rise in interest towards the SLM selection of solidstate lasers, which also promotes the development of optical components, cavity structures, and active controllers.
In a standing wave cavity, the frequency interval between each longitudinal mode (oscillation frequency) is described by where c is the speed of light in vacuum, n is the overall refractive index of the cavity, and L is the cavity length, as shown in Figure 1(a). Usually, the resonator contains a series of discrete longitudinal modes within the range of the gain curve above the threshold. at is, in the case of a freerunning laser (without SLM selection), a series of longitudinal modes with an interval of Δ] could oscillate in the spontaneous emission spectrum of the gain medium, as shown in Figure 1(b). In addition, spatial hole burning (SHB) is also a cause of multiple longitudinal modes in the cavity. e SHB effect, which results from the stronger gain saturation at locations with higher laser intensities, leads to a spatial pattern of excitation density [11]. Owing to the SHB, different longitudinal modes can use activated particles in different spaces to generate oscillations at the same time.
At present, controlling the threshold and introducing mode competition are the main approaches to generate SLM laser output. erefore, special cavity design or the introduction of mode selection components has to be carried out to generate SLM lasers. e common approaches to realize SLM operation for solid-state lasers include short cavity, intracavity etalon, unidirectional ring cavity, twisted-mode cavity, and seed injection [12][13][14][15][16][17][18]. Although many technical methods have been reported, no paper has yet provided an overview of the various SLM selection techniques and characteristics (viz. precision, cost, output power, ease of fabrication, etc.) of different methods. is paper summarizes the SLM selection techniques and reviews the developments of some typical SLM lasers, including the future prospects of the lasers in various applications.

SLM Selection Technologies
2.1. Short Cavity Method. According to equation (1), the longitudinal mode interval ∆] is inversely proportional to the optical length nL of the resonator. erefore, by shortening the cavity length, the frequency interval between adjacent longitudinal modes can be made larger than that of the gain linewidth of the medium. As shown in Figure 2, only one longitudinal mode in the gain curve reaches the oscillation threshold. is SLM selection method is called the short cavity method. To realize a large enough longitudinal mode interval via the short cavity method, a millimeter-scale cavity length is usually required for a solid-state laser.
In 1979, Kubodera and Otsuka [19] proposed a laser diode (LD) end-pumped microchip LiNdP 4 O 12 SLM laser with a cavity length of only 300 μm. is experimental setup is shown in Figure 3. Stable continuous-wave (CW) SLM outputs of 2 mW (1.048 μm) and 0.5 mW (1.317 μm) were obtained. In 2010, Li et al. [20] proposed a Tm,Ho:YVO 4 microchip laser with a cavity length of 0.5 mm. In their study, a 1.2 W multilongitudinal mode output was obtained at a cryogenic temperature of 77 K. e laser produced a single-frequency 8 mW output at 2052.6 nm when the temperature was increased to room temperature (15°C). In 2010, Wang et al. [21]  As there is no need to insert any mode selector in the cavity, SLM solid-state lasers based on short cavities are developing rapidly since the invention of the laser. However,  the reduced geometric cavity length and gain medium size of such microchip structures lead to low output power (∼mW). Meanwhile, the limited size makes it difficult to add nonlinear optical elements or modulation devices into the cavity to adjust the output characteristics of the output beam.

Intracavity Etalon Method.
Fabry-Perot (FP) etalon, based on the interference mode effect, can be used as the longitudinal mode selection element [24][25][26]. When an FP etalon is inserted, only the longitudinal mode with the highest transmittance oscillates in the cavity, while the other modes cannot reach the threshold because of the extremely low transmittance.
at is, the FP etalon increases the net gain difference between different longitudinal modes (via free spectral range (FSR)) to achieve the SLM output. e principle of FP etalon mode selection is shown in Figure 4.
As early as 1963, Collins and White [27] used a FP etalon as the mode selector, which was composed of 28.6 mm × 3.2 mm quartz flats with reflectivity of 70, 85, and 93%.
ey successfully realized single-frequency output while reducing the beam divergence angle at the diffraction limit. In 1970, Danielmeyer [28] [30], Er: YAG [31], and Ho:LuAG [32]. e etalon-based SLM laser has the characteristics of a simple overall structure, high compactness, and customized wavelength (for example, the optical material, thickness, and reflectivity of the etalon can be designed according to the cavity property). However, due to the additional intracavity loss induced by the etalon and limited threshold gap between adjacent longitudinal modes, it is difficult to generate highpower SLM output directly. erefore, this technique is often accompanied by power amplifiers when higher power is required for specific applications.

Traveling Wave Cavity.
In a homogeneously broadened laser, the mode competition caused by the gain saturation effect helps to produce SLM oscillation. However, because of the SHB effect caused by the standing wave, the laser inevitably generates multiple-longitudinal modes in the output when the pump power is sufficiently high. To eliminate the SHB effect, the beam can be designed to propagate along a single direction in the resonator, forming a traveling wave resonant cavity. Based on this idea, different structures have been developed for SLM generation including a nonplanar ring oscillator (NPRO), discrete ring cavity, and twisted mode cavity.

NPRO.
NPRO is consisting of a single laser crystal within which the laser circulates [17,33]. e front face of the laser crystal has a dielectric coating (see Figure 5), which serves as the output coupler and also as a partially polarizing element, facilitating unidirectional oscillation. In addition, all the other internal surfaces facilitate total reflection. With the action of an external magnetic field, unidirectional operation along one single polarization can be obtained easily by suppressing other polarizations, thus avoiding any standing-wave patterns that cause SHB.
In 1985, Kane and Byer [17] took the lead in realizing the LD-pumped Nd:YAG monolithic nonplanar ring cavity laser. A single-frequency laser output of 163 mW at 1064 nm was obtained. en, they improved the frequency stability by using a similar cavity type and structure, generating a singlefrequency-stabilized laser with a frequency fluctuation of 40 kHz [34]. In 2013, Wang et al. [35] demonstrated an 8.0 W single-frequency laser operating at 2.1 μm from an Ho: YAG-based NPRO. e experimental setup is shown in Figure 6. In the past two decades, different crystals have also been applied in NPRO lasers for single-frequency generation, including Tm:YAG (2.0 μm emission) [36], Er:YAG (1.6 μm emission) [37], and Nd:YAG (1.1 μm and 1.3 μm emissions) [38,39].
NPRO lasers are highly compact with high stability offering low loss which can be used to generate high-power SLM output. In addition, slow and fast tuning of the laser frequency can be realized by controlling the crystal temperature and piezoelectric ceramic of NPRO lasers. erefore, SLM lasers based on NPRO show good application prospects in the field of coherent optical communication. International Journal of Optics

Discrete Ring Cavity (Unidirectional Operation).
In recent years, ring resonance cavities based on discrete components have been developed for generating SLM output as well. Unidirectional operation is most commonly implemented through the use of an intracavity Faraday rotator. In 1972, Clobes and Brienza [40] achieved single-frequency operation of a Brewster-cut Nd: YAG laser by using a ring cavity configuration containing a small differential loss based on the use of a Faraday rotator. In 2009, Zhao et al. [41] used a four-mirror ring cavity generating 13.6 W single-frequency output with Nd:YVO 4 /YVO 4 bonded crystals as the gain medium. In the same year, Shardlow and Damzen [42] reported a single-frequency output of 17 W by constructing a threemirror ring cavity in Nd:YVO 4 slat crystals. In 2010, Zhao et al. [43] developed a 12 W single-frequency 1064 nm laser by using a four-mirror ring cavity with a conversion efficiency up to 52.7%.
Different from NPROs, this free-space traveling wave cavity structure is also conducive for the insertion of nonlinear crystals to achieve efficient intracavity nonlinear frequency conversion, thereby extending the wavelength range of SLM lasers. In addition, the frequency doubling process performed in the traveling wave cavity helps the fundamental frequency light to achieve stable single-frequency operation. e introduced nonlinear losses can suppress the inactive longitudinal mode of the fundamental frequency light in the cavity, thereby reducing the possibility of multilongitudinal mode oscillation or mode jumping [44]. In 2011, Liu et al. [45] demonstrated a CW single-frequency 532 nm laser with a six-mirror ring resonator and an intracavity frequency doubler. In their study, an etalon was inserted into the cavity to narrow the gain spectra and suppress mode hopping. e experimental setup is shown in Figure 7. In 2013, Wang et al. [46] developed a 25.3 W intracavity frequency doubling singlefrequency green laser in a four-mirror ring cavity, combined with controlling the boundary temperature.
SLM lasers based on unidirectional operation discrete ring cavities have the advantages of flexible cavity design and pump structure, which is beneficial for achieving higher output power over a wide wavelength range.

Twisted Mode Cavity.
e twisted mode cavity is another common method used to eliminate the SHB effect. e SLM laser operation is realized by inserting a polarizer into the cavity, and two quarter-wave plates at the two ends of the gain medium, respectively. In this way, the light field in the gain medium is no longer in the standing wave mode, thus eliminating the SHB effect. A schematic of the twisted mode cavity setup is shown in Figure 8.
In  [49] designed anLD-pumped SLM Pr:YLF laser that directly oscillates at 640 nm. e diagram is shown in Figure 9. Under 3.5 W blue pumping, 403 mW SLM operation with a linewidth of 150 MHz and a slope efficiency of 26.8% was obtained.
SLM laser based on twisted mode cavity is sensitive to the intracavity polarization state, so it is greatly affected by the thermally-induced birefringence of the gain medium, which limits the power scaling.

Volume Bragg Grating (VBG)
Method. VBG with high diffraction efficiency is an alternative solution to realize SLM lasing. Any wavelength change that violates the Bragg condition will result in a significant drop in diffraction efficiency.
erefore, the VBG can be used as a "filter" to achieve high reflection at a specific wavelength or band and shows good spectral and angular selectivity. Both reflective and transmissive VBGs can be used for mode selection. e working principle of the VBG is shown in Figure 10.
In 2013, Sun et al. [50] useda reflective volume Bragg grating (RBG) as the output mirror and used active Q-switching to obtain a stable 1063.9 nm SLM pulse output. e experimental setup is shown in Figure 11, and the optical cavity length of the device was 38 mm. e repetition frequency after Q-switching was 5-150 Hz, the peak power was 0.66 MW, and the pulse width was 645 ps. When the optical length reaches 100 mm, multiple modes will be generated.
e gain crystal length is only 2.6 mm, which limits the further increase in power. In 2015, Jin et al. [51] used RBG and graphene passive Q-switching to obtain a single-frequency laser pulse with an output power of 724 mW, single pulse energy of 7.5 μJ, repetition frequency of 96.2 kHz, and duration of 2.2 μs. Passive Q-switching increases the pulse set-up time, which is conducive to the realization of a SLM. However, the passive Q-switched output pulse has a time jitter, which limits its application range.
Owing to the limitation of the gain medium of the laser, the length of the resonant cavity cannot be further shortened. is may cause the grating bandwidth to be longer than the longitudinal mode interval. It is impossible to achieve SLM operation with a single VBG. erefore, multiple VBG are required, and mode selection is carried out by a combination of these gratings. In 2009, Hui et al. [52] proposed a SLM laser based on a combined VBG, and its  e Bragg grating is limited by the accuracy of the mode selection and is generally used as a part of the composite mode selection structure. e RBG is usually used as the output mirror of the resonant cavity.

Seed Injection and Amplification Method.
e output powers of the abovementioned SLM selection techniques in a single cavity are relatively low. To generate higher power output, the seed injection and amplification method is proposed. e basic principle is as follows. A single-frequency low-power seed with excellent temporal and spatial characteristics is injected into the resonator. e cavity length continuously changes near the descending edge of the pumping pulse and detects the match of the seed light and the oscillation mode in the cavity. In this case, when the seed light resonates in the cavity, the Q switch is turned on, thereby outputting a single-frequency pulse laser with a linewidth near the Fourier transform limit. Park et al. [53] showed that the frequency of the SLM output by the seed injection solidstate laser is not exactly the same as the seed light frequency, but is the longitudinal mode frequency closest to the seed light frequency. Seed injection technology plays a vital role in the energy stability and frequency stability of the output laser and is the core technology of singlefrequency lasers. Commonly used seed injection techniques include minimum setup time, resonance detection technology, and optimized resonance detection technology. A schematic of this technology is shown in Figure 13.

International Journal of Optics
In 2007, Schröder et al. [54] designed a seed-injected Nd: YAG single-frequency ultraviolet laser for airborne Doppler wind lidar using the technology of minimizing the settling time. A 100 Hz pulse with a duration of 35 ns and a line width measured by beat frequency to be less than 15 MHz was passed through a two-stage dual-pass amplifier and nonlinear crystal.
is resulted in an output of 355 nm ultraviolet laser with pulse energy of 60 mJ and pulse width reduced to 25 ns.
In 1986, Henderson et al. [55] proposed a scan-triggered resonance detection technology. In 2017, Gao's research group developed a seed-injected Ho:YAG single-frequency laser [56], using a fiber with a line width of approximately 10 kHz. e coupled output Ho:YAG monolithic nonplanar ring cavity (NPRO) laser is used as the seed source, and the driven cavity is an acousto-optic Q-switched butterfly fourmirror ring cavity pumped by a continuous-wave Tm:YLF laser. e structure is shown in Figure 14. Ramp-fire resonance detection technology was used to achieve a stable 2.09 μm SLM laser output with output pulse energy of 6.24 mJ, pulse width of 172 ns, and repetition frequency of 1 kHz. e spectral linewidth was measured as 2.61 MHz using the optical beat frequency method, and the energy stability was less than 3%. e energy stability index has room for improvement.
To solve the problem of output laser energy jitter caused by the unstable Q-switched trigger signal of the Ramp-fire resonance detection technology, researchers have proposed several improvements. In 1997, Fry et al. [57] proposed a scan-hold-fire (Ramp-hold-fire) resonance detection technology based on Ramp-fire resonance detection technology. e principle of this technology is similar to Ramp-fire resonance detection technology. e difference is that when the interference peak is detected, the Q switch is not turned on immediately, but the scanning voltage applied to the piezoelectric ceramic is maintained at the same level as that at the time when the interference peak is detected. At this time, the longitudinal mode frequency of the driven cavity remains consistent with the seed light frequency until the Qswitch trigger signal arrives. e technical working principle is shown in Figure 15.
In 2016, Zhang et al. [59] developed the first seed-injected Ho:YAG ceramic laser using this technology. e laser used a 2.09 μm seed laser with an output pulse energy of 14.76 mJ, pulse width of 121.6 ns, and repetition frequency of 200 Hz. e linewidth was 3.84 MHz for one-hour operation.
In order to further improve the stability of the output laser energy and avoid the influence of external interference as much as possible, Ertel et al. [60] proposed a resonant detection technology with feedback in 2005. In 2007, Zhou et al. [61] proposed a delay-sweep-trigger (delay-ramp-fire) resonance detection technology and an improved scheme of this technology, the double piezoelectric ceramics (PZT) resonance detection technology with bias feedback [62]. In 2016, Gao et al. [63] combined the real-time resonance tracking detection method and proposed a new seed injection technology: a double PZT sinusoidal scanning resonance detection technology with bias voltage feedback. e singlefrequency laser developed by this technology has an output pulse energy of 0.7 mJ, a pulse width of 27 ns, a repetition frequency of 400 Hz, and a line width of 18 MHz. A frequency jitter of less than 9.1 MHz was measured within 30 mins. In 2014, Gibert et al. [64] used Pound-Drever-Hall (PDH) injection locking technology to develop a dual-wavelength single-frequency Ho:YLF laser. e standard deviation of the output laser frequency stability was 2 MHz, and the frequency jitter range within 10 s was less than 70 kHz, which is very suitable for differential absorption radar to detect atmospheric CO 2 concentration. In 2017, in the German Aerospace Center Lemmerz et al. [65] adopted the scan-delaytrigger (Ramp-delay-fire, RDF) resonance detection technology to develop an all-solid-state ultraviolet laser for airborne wind measurement radar. e oscillation amplification structure, combined with type I LBO frequency double   Figure 12: SLM laser based on combined VBGs [52].
International Journal of Optics crystal and type II LBO sum frequency crystal, achieved an ultraviolet laser with an output pulse energy of 60 mJ, pulse duration of 20 ns, and repetition frequency of 355 nm at 50 Hz. Figure 16 shows the principle of the trigger mechanism of the RDF resonance detection technology. T 1 is the pump start time, T 2 is the PZT sweep voltage start time, peak detection starts at a fixed time T 3 , and the first peak is detected at T 4 . In order to trigger the radar return light detector at a precise time before the pulse is generated, the trigger time T 6 of the next pulse is predicted based on the difference between T 3 and T 4 to determine the time T 5 to trigger the return light detector, and the cavity is changed within the delay time D 2 to compensate for the change in the cavity length caused by the PZT during this period (including the pulse setup time). is method can reduce the pulse time jitter caused by the external interference in the ramp-fire technology and help in finally realizing a single-frequency laser emission source with stable frequency and time. e seed injection method can effectively increase the output power of the laser, and the beam quality of the seed laser has a great influence on it. erefore, it is very important to use a suitable seed laser injection technology. In current technologies, the seed injection method is often cascaded with nonlinear crystals to form a frequencydoubled high-power laser [66].

Compound Cavity Method.
Compared to a resonant cavity with only one light propagation path, the compound cavity has two or more optical transmission paths that have common parts. Different optical transmission resonators are called subcavities, and a compound cavity consists of multiple subcavities. Depending on whether the light propagation path is separated in the direction, the compound cavity can be divided into the same direction as the compound cavity and the bifurcation composite cavity. A typical compound cavity mode selection structure is shown in Figure 17. e mirror at one end of the laser is replaced by a combination of three mirrors. Among them, M 3 and M 4 are total reflection mirrors and M 2 is a mirror with appropriate transmittance. For transmission mirrors, this combination is equivalent to the coupling of two resonant cavities: one resonant cavity is composed of M 1 and M 3 with a cavity length of L 1 + L 2 , and the other resonant cavity is composed of M 3 and M 4 with a cavity length of L 3 + L 2 . If L 2 and L 3 are short, a coupling of a short resonant cavity and a long resonant cavity will be formed. e longitudinal mode frequency interval of the short resonant cavity is e longitudinal mode frequency interval of the long cavity is where μ is the refractive index of the gas in the cavity. e light oscillates only when the above two resonance conditions are reached. erefore, as long as L 3 + L 2 is selected to be sufficiently small, a SLM output can be obtained. e interferometer compound cavity is usually used to generate a SLM laser. e Michelson interference ceremonial compound cavity [67] and the Fox--Smith interference ceremonial compound cavity [68] are two representatives of the interference ceremonial compound cavity. Among allfiber lasers, there are many examples of compound cavities realized by optical fibers, and there are fewer applications of compound cavities in solid-state lasers. In 2016, Bai et al. [69] demonstrated an SLM Q-switched laser based on a three-plan resonant reflector, which is a linear cavity compound cavity. e laser delivers single-pulse energy of 10 mJ and a pulse width of 10.7 ns at 10 Hz.

Nonlinear Frequency Conversion Method.
Nonlinear frequency conversions such as frequency doubling, stimulated Raman scattering, and stimulated Brillouin scattering provide a valuable additional mechanism for increasing gain competition and therefore enhancing SLM stability [44,70,71]. For example, SLM laser operation can be obtained in simple cavities even without single-mode constraints on the pump laser owing to the SHB free and homogeneous nature of Raman gain.  Figure 16: Schematic of the trigger mechanism of RDF resonance detection technology [65].  Figure 18. In their study, multi-watt, stable SLM Stokes (∼1178 nm) was derived from the multimode Nd:GdVO4 fundamental (linewidth <0.1 nm @1.06 μm) through hole burning free Raman gain. en, SLM yellow output at the wavelength of 589.16 nm is obtained by intracavity frequency-doubling of the Stokes wave when it was tuned to 1178.32 nm.
In 2019, Yang et al. [73] developed a single-frequency diamond Raman laser with output powers of 38 W at 620 nm and 11.8 W at 1240 nm. e Raman cavity is based on a simple standing-wave structure pumped by a multilongitudinal mode laser. Subsequently, the same group [5] demonstrated a SLM 589 nm laser with near-diffraction limited beam quality generated in a standing-wave diamond Raman resonator with intracavity second-harmonic generation. With a 63 W 1018 nm pump, 589 nm laser with output power up to 22 W was achieved. e experimental configuration is shown in Figure 19.
SLM laser operation based on nonlinear frequency conversion can lead to more kinds of frequency lasers. However, this method has a high requirement of nonlinear crystals, and its output mode quality is limited by the material quality and the nonlinear properties of the crystal.

Multistructure
Combination. In practice, due to the processing technology of the parts and the characteristics of the structure itself, a single-mode selection structure is increasingly unable to meet the requirements for SLM selection. To overcome this, multiple structures have been combined to improve the SLM selection level. In 2014, Zhang et al. [74] of the Shanghai Institute of Optics and Mechanics used the torsion cavity structure, combined with the resonance detection technology, to obtain the doublepulse SLM pulse output, as shown in Figure 20. When using dual-crystal RTP electrooptical modulation, the single pulse energy could reach 10 mJ. e pulse width of the generated output was 20 ns, the repetition frequency was 50 Hz, and the optical conversion efficiency and slope efficiency were 22.4% and 33%, respectively. However, if PZT control the cavity length to meet the resonance conditions, fluctuations will occur, and the hysteresis effect of PZT will reduce the stability. In order to overcome the shortcomings of PZT, Zhang et al. [75] replaced PZT with electrooptic crystals in 2017.
e scanning-delay-triggered resonance detection technology was used to control the cavity length, which improved the stability of the laser. e pulse energy was increased to 13 mJ, the pulse width was 20 ns, the repetition frequency was 50 Hz, and the optical conversion efficiency and slope efficiency were 27% and 34%, respectively.
In 2019, Hu et al. [76] and others used the RBG and FP etalons to jointly select the mode and used the pressurized Q-switch method. e structure is shown in Figure 21. A stable single frequency could be achieved with a cavity length of 145.7 mm. A pulse of output power of 750 mW, single pulse energy of 75 μJ, frequency 10 kHz, and pulse width of 8.3 ns could be obtained using this technique.
In 2020, Jin et al. [77] proposed a structure in which active and passive dual Q-switching are combined with multiple longitudinal mode selection technologies. is structure is shown in Figure 22. Using a saturable absorption crystal to adjust the cavity loss, combined with a resonant reflector (etalon) and twisted-mode technology, a symmetrical pulse laser output with a single pulse energy of 10.8 mJ, pulse width of 9.8 ns, and an SLM rate of 96.2% was obtained.

Summary and Outlook
is review provided an overview of the main SLM selection methods used in lasers, mainly the short cavity method, the etalon method, the traveling wave resonator method, the VBG method, the seed injection method, the compound cavity method, and the nonlinear crystal method. e advantages and disadvantages of different methods are compared in Table 1.
With the advancement of scientific research, driven by the increasing practical demand, the performance of SLM lasers will become progressively better in the future. Miniaturized lasers with high stability, high efficiency, narrow line width, and large energy will become a reality in the future. e seed injection method, nonlinear crystal method, and compound structure method are the frontrunners in future selection technologies. e advantage of the high power of the seed injection method, the nonlinear frequency conversion characteristics of the nonlinear crystal, and the high precision of the composite structure are important factors in practical applications. In addition,   Figure 19: Schematic of the 589 nm diamond Raman laser with intracavity second-harmonic generation [73]. the use of fiber resonators in solid-state lasers may also help in making the use of composite resonators more widespread.

Conflicts of Interest
e authors declare that there are no conflicts of interest regarding the publication of this paper.