1. Introduction

There is a rapidly growing demand for high-power Cr²⁺:ZnS and Cr²⁺:ZnSe lasers operating in the mid-IR spectral range of 1900–3400 nm [1, 2] for numerous applications in the fields of industrial material processing (e.g. fast cutting, welding, marking of polymers near 2400 nm wavelength region, processing of glasses in the range of 2700–2950 nm), laser surgery near 2940 nm, and science. These promising laser optical material were introduced to the scientific community as early as 1996 by Lawrence Livermore National Laboratory [3, 4]. Since then, and noticeably in recent years, many Cr²⁺:ZnS and Cr²⁺:ZnSe laser systems have been demonstrated covering a broad range of regimes of operation and output parameters [5–11]. However, in order to be successful in the industrial applications, where high throughput is critical, mid-IR laser systems must deliver tens and hundreds of Watts of output power with very high beam quality and overall system reliability.

Availability of high-power fiber lasers operating in the 1.5–1.9 μm spectral range and our core technology of manufacturing high-quality large size Cr²⁺:ZnS and Cr²⁺:ZnSe gain elements have resulted in significant progress in the development of practical mid-IR laser systems based on these laser gain media [12]. However, due to the inherent properties of these optical materials, the maximum output power with practically acceptable beam quality was limited to approximately 10 W for 2.4–2.5 μm lasers and 2 W for 2.94 μm laser systems [13]. The main limiting factors for obtaining high output power in Cr²⁺:ZnS and Cr²⁺:ZnSe lasers based on either conventional slab gain media or thin disk geometries are very strong thermal lensing effects [14, 15] and relatively low optical damage threshold [16, 17]. To manage thermal effects of thin disks one has to use small disk thickness which results in small optical density, thus mitigating efforts of power scaling.

The novel approach to significantly resolve these challenges, which we demonstrate in this work, consists of performing a simultaneous rapid infinite transverse scan of the coaxial laser mode and pump beam across the gain element along a closed or open path. Here we describe our proof of concept experiments and demonstrate previously unseen levels of output powers from Cr²⁺:ZnSe CW laser systems. We experimentally consider several laser configurations and resonator designs, including standalone master oscillators and power amplifier systems (MOPA). We also demonstrate various pump configurations based on single- and multi-pump sources.

2. Technology background and first proof of concept experiments

In recent years we have developed and delivered a broad range of high-power CW laser systems based on Cr²⁺:ZnS and Cr²⁺:ZnSe gain media, which include but are not limited to: widely-tunable single-frequency lasers with a maximum output power of up to 5 W and a tuning range of up to 550 nm; narrow-line widely tunable lasers with typical linewidths of <0.25 nm, tuning ranges well over 1000 nm, and maximum output power reaching 10 W level; multiple models of fixed-wavelength lasers both in single-frequency and narrow-line configurations with multi-Watt output power [18]. To further boost the output power of these laser systems we explored various configurations based on resonator designs with multiple gain elements and multistage power amplifiers and obtained some promising preliminary results [19].

Unfortunately, due to relatively high thermo-optic coefficients in host ZnS and ZnSe materials [20], obtaining tens of Watts output power in conventional slab or rod gain element geometries have proved to be very difficult: at high levels of pump power thermal lensing rapidly leads to significant mode distortion as the pump power is increased, which results in rapid heat buildup in the active region, significant degradation of laser efficiency due to growing non-radiative relaxation, unusable output beam quality, and eventually optical damage of the gain element. Based on numerous experiments and tests of our standard and experimental laser systems with a large variety of gain element optical densities and sizes we found that ∼10 W to ∼15 W of absorbed pump power per gain element is the upper acceptable limit for all practical purposes (the exact limit depends on pump wavelength, mid-IR laser type, output wavelength and tuning range). These difficulties called for active exploration of alternative methods and laser designs for power scaling of Cr²⁺:ZnS and Cr²⁺:ZnSe based laser systems.

Our novel approach to virtually eliminate thermal lensing effects in the Cr²⁺:ZnS and Cr²⁺:ZnSe gain media is based on a simple observation that the strength of the thermal lens is determined by the total thermal energy absorbed by the volume of the gain element being pumped. Indeed, if the heat energy absorbed by the active region of the gain element is sufficiently small, the local refractive index change is insignificant, and the thermal lens is so weak that it does not have any noticeable effect on laser performance. On the other hand, the total absorbed thermal energy is directly proportional to the average power of the pump radiation. This observation is clearly supported by numerous experiments with QCW lasers operating at low duty cycles as well as gain-switch lasers where we observe hundreds of Watts, kW or even MW peak power levels with very high optical to optical efficiencies and high beam quality [21].

The proposed method to reduce heating of the active region of the Cr²⁺:ZnS or Cr²⁺:ZnSe gain element is to perform a simultaneous continuous rapid scan of the collinear pump beam and laser mode across the gain crystal with a sufficiently high speed to keep the instantly absorbed pump power below a certain level. In this case we expect to obtain highly-efficient CW lasing with any level of output power while maintaining good beam quality.

In order to investigate the feasibility of this approach and test out proposition, we first conducted a simple experiment with the laser setup shown schematically in Fig. 1. In this laser setup, a conventional slab Cr²⁺:ZnSe AR-coated gain element is mounted on a water-cooled copper heatsink and we scan the collinear pump beam and laser mode with the help of an octagon mirror.

 

Fig. 1 Design of proof-of-concept laser resonator. The laser resonator is formed by the end mirror, cavity/pump lens, dichroic mirror, and output coupler (OC). The pump beam and laser mode are collinear and coupled through the dichroic mirror (AR@1908 nm, HR@2400–3000 nm). The spinning octagon mirror and fixed angular retro-reflector scan the laser mode and the pump beam across the AR-coated Cr²⁺:ZnSe gain element at normal incidence. Two octagon angular positions, corresponding to extreme locations of the laser mode and pump beam in the gain element, are shown.

Download Full Size | PDF

Due to the relatively high intracavity total losses caused by double-reflections from machined aluminum surfaces of the octagon and gold mirrors of the angular retroreflector, the laser has relatively low efficiency (which in fact mimics Cr²⁺:ZnSe laser performance near 2940 nm and other challenging spectral regions where the gain is relatively low). Nevertheless, this experiment verifies the premise of our approach and allows us to practically investigate its potential, limitations, and requirements on the parameters of a practical laser system. The performance comparison of this laser in pure CW regime of operation (scanner is OFF) and QCW mode (scanner is ON) is illustrated by the input-output characteristics shown in Fig. 2.

 

Fig. 2 Input-Output characteristics of pure CW (scanner is OFF) and QCW (scanners is ON) regimes of operation of the laser system shown in Fig. 1. In QCW regime of operation the duty cycle is ∼9%, linear scan speed is ∼1 m/s. The graph insert shows QCW trace recorded with a fast optical detector.

Download Full Size | PDF

One can see from Fig. 2 that in pure CW regime of operation, corresponding to the conventional gain element slab geometry, the laser demonstrates a rapid power rollover just above 15 W of pump power. Further pump increase leads to a complete laser shutoff and eventually results in surface and internal optical damage of the gain element.

In sharp contrast, when the pump beam is rapidly scanned across the gain element (QCW regime in this case), the laser accepts much higher peak pump power and generates much higher instant output power. Consequentially, if we manage to construct a system with an effectively infinite and uninterrupted scan of the pump beam across the gain element we can expect to obtain very high pure CW powers limited only by the available pump sources.

3. Spinning ring gain element approach

In order to implement this idea and thoroughly investigate the limitations and capabilities of this approach we manufactured Cr²⁺:ZnSe and Cr²⁺:ZnS gain elements in the form of closed rings. The ring is installed on the shaft of a motor and is rotated at a high speed, thus providing an effectively infinite scan of the collinear laser mode and pump beam across the gain element. The test spinning ring system design is shown schematically in Fig. 3. This approach is not equivalent to, but to some extent similar to prior publications devoted to laser systems based on rotary disk gain elements [22–24].

 

Fig. 3 Test design of the spinning ring gain element system: (A) Technical drawing (front and side views) of the Cr²⁺:ZnSe spinning ring gain element with key dimensions. The Cr dopant concentration is 5 × 10¹⁸ cm⁻³, gain element thickness is 7 mm. The ring gain element has been manufactured with the following specifications and tolerances: (1) Scratch-Dig: 40/20; (2) Flatness: 1/6 wave at 632 nm; (3) Wedge: <10 arc sec; (4) Concentricity error between the external and internal circumferences: <0.05 mm; (5) AR coating specifications (measured per surface): R<0.2% @1.9 μm, R<0.8% @1.9–2.8 μm, R<0.5%@2.94 μm. (B) Expanded view of the test opto-mechanical system. The spinning ring is mounted between 2 cooling flanges. Indium foil (not shown in the scheme) is used to provide improved thermal contact between the gain element ring and flanges. The spinning flanges are self-cooled in open air at moderate heat loads (up to 100 W of total incident pump power). The flanges are additionally cooled with 2 compressed air jets directed from one side. The ring is spinning at a nominal speed of ∼9500 RPM. (C) Side broken view if the test opto-mechanical system also showing additional cooling air jets and numbering scheme of the cooling fins for thermal analysis. The fins of cooling flanges are numbered with integers, the grooves are numbered with fractions, numbers −0.5, 0, 0.5 designate the locations at the Cr²⁺:ZnSe gain element where surface temperature was measured. The pump radiation enters the spinning ring from direction of fin −6 (left in this drawing).

Download Full Size | PDF

In the framework of this investigation we have conducted a substantial set of experiments with various configurations of laser resonators utilizing the spinning ring gain element approach. We have tested the performance of standalone master oscillators in free-running mode with broadband laser mirrors, tunable laser sources, and a number of systems with wavelength selection optical elements. We have also conducted a thorough investigation of spinning ring gain element performance in power amplifiers.

In this work we show only a selected subset of laser resonator designs and experimental results which represent the most promising laser configurations suitable for practical applications. Figure 4 shows by far the simplest MOPA design based on spinning ring gain element technology with the highest optical efficiency. It is based on a very simple 2-mirror design of the master oscillator, single-pass power amplifier, and a dual pump source configuration. Figure 5 shows a standalone master oscillator design for a high-power 2940 nm CW laser system intended for industrial applications. Figure 6 shows a laser resonator design with dual pump sources.

 

Fig. 4 Schematic diagram of the test MOPA laser system based on the spinning ring gain element approach (relative scale is preserved, optomechanical mounts are not shown). The laser system consists of a simple master oscillator and a single-pass power amplifier. The system is pumped with two 100 W Tm-fiber lasers (IPG TLR-100-1908-WC models). The output wavelength is generally determined by the spectral curve of the output coupler. In the range of 1950–2400 nm, narrow-linewidth radiation (δλ ≤ 0.25 nm) is obtained by spectral control with volumetric Bragg gratings (VBG).

Download Full Size | PDF

 

Fig. 5 Schematic diagram of high-power 2940 nm laser system based on the spinning ring gain element laser technology. Wavelength selection is performed by the output coupler and a triple dichroic mirror set (input mirror and 2 intracavity mirrors) which force the laser to oscillate near 2940 nm. Due to relatively low gain of Cr²⁺:ZnS and Cr²⁺:ZnSe laser media near 3 μm, strict wavelength control is essential to prevent the laser from free-running oscillation near its gain maximum within the range of 2300–2500 nm.

Download Full Size | PDF

 

Fig. 6 Schematic diagram of a laser resonator design based on the spinning ring gain element laser technology with dual pump source. This laser system can also be used as the basis for high power tunable lasers when the end mirror is replaced with diffraction grating in the Littrow or Littman mount configurations. A tuning prism or Liot filters can also be used for wavelength control within a limited spectral range.

Download Full Size | PDF

It is noteworthy that all cavity configurations used off-the-shelf 50 mm FFL (Front Focal Length) CaF₂ lenses with custom AR coatings spanning 1.5–3.1 μm. In some laser configurations the same lens is used for pump focusing and also as a laser resonator component, which is very attractive for reasons of system design simplicity and cost-efficiency. The Tm-fiber laser pump source operating at central wavelength λ = 1908 nm emits a collimated laser beam with a low divergence and beam diameter of 2w_p ≈ 1.2 mm. The pump lens with a focal length F = 50 mm focuses the pump laser beam and creates a beam waist whose diameter can be estimated as 2w₀ ≈ 4Fλ/2w_pπ ≈ 101 μm. Using the (ABCD)–matrix formalism, one can find that in the simple single-lens master oscillator cavity the laser mode waist can be matched with the pump beam waist when the intracavity lens is located at approximately 62 mm from the output facet of the gain element (the lens to the OC distance is ∼ 150 mm). In practice, we set the lens at this theoretically calculated location and then gradually adjust its axial position to obtain the lowest possible lasing threshold. Our experiments show that due to efficient suppression of thermal lensing effects in the gain media, the minimum lasing threshold cavity adjustment is close to optimal even at the highest pump power levels.

The more complex 4-lens laser resonators and the single-pass power amplifier are theoretically consist of single-lens resonator “chains” and thus should not require any significant modifications for keeping good mode matching conditions. In practice, however, some minor optimization of axial positions of all lenses is required to maximize optical efficiency.

4. Major experimental results

In this section we present our most significant results obtained with Cr²⁺:ZnSe laser systems based on the spinning ring gain element technology. We have tested the performance of these laser systems in single master oscillator configurations, single-pass MOPA arrangements, dual-pump laser systems, and beam combined laser designs.

An important note must be made here regarding the selection of pump sources for Cr²⁺:ZnS and Cr²⁺:ZnSe laser gain media. In all our CW mid-IR laser systems we utilize collinear pumping approach (also known as “end pumping”). This approach provides the highest possible optical to optical efficiency when there is a good mode matching between the pump beam and the laser mode. The latter is readily achievable when the pump source delivers high beam quality with M² factor close to 1.0, which is why we choose high-power fiber lasers with a single-mode delivery fiber optics for pumping.

There are generally 2 families of commercially available high power fiber lasers suitable for pumping Cr²⁺:ZnS and Cr²⁺:ZnSe laser systems: Er fiber lasers operating near 1.5 μm and Tm fiber lasers with output wavelengths near 1.9 μm. Currently, Tm fiber lasers deliver much higher output powers than Er fiber systems. Since the main goal of this work was to perform power scaling of the mid-IR laser to the highest possible level, we chose to use Tm-fiber lasers as the optical pump sources for this work.

It is also noteworthy that we have conducted preliminary power-scaling experiments with Cr²⁺:ZnS spinning ring gain media and obtained very promising results comparable to the performance of spinning ring Cr²⁺:ZnSe lases. Unfortunately, the AR coating of Cr²⁺:ZnS spinning rings was not suitable for pumping at 1.9 μm and we decided to devote a separate publication to this gain material in the future.

4.1. Master oscillator performance

Figure 7 shows general performance of the master oscillator configured for different output wavelengths. The output wavelength of 2500 nm results from using broadband 50% output coupler and represents a free-running CW oscillation of the master oscillator based on spinning ring Cr²⁺:ZnSe gain element. The output wavelength of 2300 nm is obtained with a volumetric Bragg grating (VBG) manufactured by Optigrate, Inc. The output wavelength of 2940 nm is determined by a 4-mirror selective laser resonator shown in Fig. 5. The output spectra of the master oscillators corresponding to these regimes of operation and laser resonator configurations are shown in Fig. 8, Fig. 9, and Fig. 10.

 

Fig. 7 Input-Output characteristics of the Cr²⁺:ZnSe master oscillator at different output wavelengths. The laser demonstrates ∼ 63% absolute and ∼ 65% slope efficiencies near 2.5 μm in free-running CW regime of operation with bradband 50% output coupler (FR); ∼ 59% absolute and ∼ 62% slope efficiencies at 2.3 μm where the central wavelength is selected with 50% efficient VBG designed for 2300 nm (VBG); and ∼ 29% absolute and ∼31% slope efficiencies near 2.94 μm, where the output spectrum is determined by the 4-mirror intracavity selector (MS). The data labels show maximum output power values (the label values are rounded to whole Watts).

Download Full Size | PDF

 

Fig. 8 Measured output spectrum of the master oscillator near central wavelength of 2500 nm. The laser is operating in free-running CW regime with a broadband output coupler (OC) and the output wavelength is determined by the location of the gain maximum of Cr²⁺:ZnSe gain media. In this free-running mode, the laser shows approximately 40 nm linewidth FWHM.

Download Full Size | PDF

 

Fig. 9 Measured output spectrum of the master oscillator at 2300 nm. The central wavelength is determined by a specific custom-made VBG output coupler with ∼ 50% reflectivity. The VBG OC results in extremely narrow output linewidth of less than 0.25 nm (which is actually the upper estimate determined by the resolution of available laser spectrum analyzer, Bristol 721).

Download Full Size | PDF

 

Fig. 10 Measured output spectrum near 2.94 μm. The output spectrum is determined by intracavity 3-mirror selector and specific output coupler in the laser resonator design shown in Fig. 5. Due to relatively broad slopes of the cut-off edges of the selector mirrors and OC dielectric coatings, the laser demonstrates relatively broad linewidth of approximately 30 nm FWHM.

Download Full Size | PDF

4.2. MOPA performance

The greatest advantage of rapidly scanning the laser mode/pump beam through gain media is the almost complete elimination of thermal lensing effects. As a result, these laser systems behave almost identically to pure theoretical laser devices where all thermal effects are generally ignored. This opens enormous opportunities for significant power scaling of these laser systems. One can see in the previous data that no thermal rollover is observed even at such unfavorable output wavelengths as 2.94 μm, where gain is very low and passive losses are very high (due, in particular, to intracavity water vapor absorption in un-sealed and un-purged cavities). This also results in excellent performance of single-pass power amplifiers, as is demonstrated in the experimental data shown in Fig. 11, where single-pass amplifier MOPA input-output characteristic at 2.3 μm is shown. Figure 12 demonstrates calculated gain as a function of input master oscillator signal when the single-pass amplifier is pumped at full 117 W pump power.

 

Fig. 11 Input-Output characteristic of MOPA system operating at 2300 nm. The laser demonstrates ∼ 55% absolute and ∼ 57% slope efficiencies at 2.3 μm

Download Full Size | PDF

 

Fig. 12 Calculated gain of single-pass power amplifier as a function of incident master oscillator power. The gain is calculated as the ratio of output power to the master oscillator incident power. The amplifier incident pump power is constant and equals approximately 117 W.

Download Full Size | PDF

4.3. High-power performance of Cr²⁺:ZnSe laser systems

In this section we present our best power scaling results wit Cr²⁺:ZnSe spinning ring laser gain media. We show that scanning the laser mode/pump beam across the gain media allows for obtaining unprecedented output powers in the 2–3 μm spectral region with Cr²⁺:ZnSe gain material. Figure 13 shows the input-output characteristic of a 140 W laser system where the output beams of two simple master oscillators shown in Fig. 4 are polarization combined as well as the more complex dual-pump laser resonator, shown in Fig. 6. Figure 14 demonstrates typical power stabilization curves after cold start of two laser systems: (1) dual-pump free-running laser operating near 2500 nm, and (2) 2.94 μm laser system. Figure 15 represents typical output beam profiles obtained with our power-scaling approach from the laser configurations discussed above.

 

Fig. 13 Performance of two beam-combined maser oscillators and 4-lens laser cavity with dual-pump source in single spinning ring laser system. The complete absence of thermal rollover demonstrates unsaturated pump power capability of these laser systems and indicates further power scaling potential for reaching kW levels of output power.

Download Full Size | PDF

 

Fig. 14 Typical power stabilization curves after cold start of high-power Cr²⁺:ZnSe laser systems based on the spinning ring gain element approach. Power stabilization of 4-lens free-running laser resonator operating near 2.5 μm and 2.94 μm master oscillators are shown.

Download Full Size | PDF

 

Fig. 15 Typical beam profiles of the spinning ring laser systems in various configurations. The beam profiles were acquired with Pyrocam III™ pyroelectric laser beam analyser located approximately 0.5 m from the laser output mirror. The camera was operated in CW laser mode.

Download Full Size | PDF

In order to perform a more accurate characterization of beam quality of Cr²⁺:ZnSe laser systems based on spinning ring gain element technology we measured M² factor of the least efficient master oscillator operating near 2.95 μm. Due to relatively low optical to optical efficiency of this laser system the influence of residual thermal lensing effects on the output beam quality is more pronounced as compared to other output wavelengths within the gain profiles of Cr²⁺:ZnS and Cr²⁺:ZnSe laser systems.

The M² measurements were performed using the Knife-Edge method (described in very details in Ref. [25]) as follows. A small fraction of the output power was sampled with a CaF₂ wedge and focused by a 100 mm FFL AR coated CaF₂ lens. A razor blade was scanned across the laser beam with a motorized linear stage moving at a constant speed of 0.05 mm/s. The transmitted signal was recorded as a function of time with the help of Ophir™ P3-A1 thermal sensor, Nova II™ power meter, and StarLab™ software. The beam radii were measured at multiple locations along the beam axis. The measurement results are shown in Fig. 16.

 

Fig. 16 Measurement results of beam quality M² of the 32 W master oscillator operating near 2.95 μm. The fitting parameter $A_0\approx w_0^2$ is the beam waist radius, the fitting parameter A₁ is the offset of w₀ from relative measurement origin, the fitting parameter A₂ is the confocal range. The value of M² ≈ 1.77 is estimated from the fitting parameter A₀ and A₂ as: $M^2\approx \pi A_0/\left(\lambda A_2^{1/2}\right)$, where λ ∼ 2.95 μm is the output wavelength.

Download Full Size | PDF

It is evident from the beam profiles shown in Fig. 15 as well as from the measured M² value of 1.77 that some residual beam distortion caused by heat build-up in the gain element is still present. An additional factor that contributes to non-ideal beam quality is the presence of acoustic vibrations due to very simple mechanical design of the spinning ring opto-mechanical subsystem. Nevertheless, the presented results clearly demonstrate feasibility and a great potential of the spinning ring gain element approach for significant power-scaling of the Cr²⁺:ZnSe laser systems. Further improvement of gain element cooling system and spinning drive mechanism will result in significant improvement of output beam quality at very high output power levels.

4.4. Thermal performance

As it was mentioned earlier in the paper, the primary goal of this research project was to investigate practical feasibility of the spinning ring gain elements for significant power scaling of Cr²⁺:ZnSe and Cr²⁺:ZnS laser systems via efficient thermal management of the active media. Experimental analysis of thermal performance of the opto-mechanical system used in these experiments is of great importance for refining major parameters of the spinning gain element and cooling arrangement. The latter is especially important within the framework of further power-scaling of Cr²⁺:ZnSe and Cr²⁺:ZnS laser systems to kW output power levels.

In order to characterize heat built-up in the gain medium and performance of the cooling subsystem we conducted direct measurements of the spinning ring gain element temperature and cooling flanges. The experimental data of these thermal measurements are currently used in development if a practical thermal model of the spinning gain element subjected to optical excitation at very high power levels. For direct temperature measurements of the spinning components we used a remote infrared thermometer (Cen-Tech™) equipped with a built-in pointing laser beam. Preliminary calibration showed that one can expect an ±2°C spot temperature measurement accuracy. The thermal measurement results are demonstrated in Fig. 17. These measurements were performed for the dual-pump laser system operating near 2500 nm with a total incident pump power of ∼ 226 W (estimated transmitted pump does not exceed 10 W).

 

Fig. 17 Temperature profiles of the cooling flanges and spinning ring gain element of dual-pump, 4-lens laser system. The x–axis indicates fin/spot numbers as shown in Fig. 3(C). The “Simulation” curve is given for illustrative purposes and corresponds to thermal profile obtained using Solidworks™ thermal simulation package. The measurements at zero pump level (green curve) show a gradual heating of the right cooling flange due to operation of the DC motor.

Download Full Size | PDF

5. Conclusions and future outlook

In this work we have demonstrated unprecedented output power levels of mid-IR laser systems based on Cr²⁺:ZnSe gain media. We have utilized a novel approach for this gain media based on scanning the collinear pump beam and laser mode across the gain element to obtain output power levels of up to 140 W with real optical efficiencies reaching ∼62% near the gain maximum (2.5 μm) and ∼ 29% at the key wavelength of 2.94 μm. The input-output characteristics of various laser configurations based on our novel approach show significant unused pump power capacity and indicate that we can potentially reach kW output power levels in the spectral range of 2000–3000 nm and beyond. These results clearly show that Cr²⁺:ZnS and Cr²⁺:ZnSe laser systems have come of age and have became suitable for practical laser systems highly demanded for material processing applications in the fields of medicine, industry, as well as cutting-edge research.

Besides simple power scaling of CW mid-IR lases based on Cr²⁺:ZnS and Cr²⁺:ZnSe gain media there is a very promising possibility of significant power-scaling of mode-locked systems based on the same gain materials [26] using the spinning ring gain element approach. The simplest way to boost the output power of mode-locked Cr²⁺:ZnS and Cr²⁺:ZnSe lasers is to utilize a single-pass spinning ring power amplifier. According to Fig. 12, the output powers in excess of 80 W should be readily available with the use of combination of 5–7 W Cr²⁺:ZnS femtosecond seed (described in [26]) and a single-pass spinning ring amplifier pumped by 100–120 W radiation of Tm-fiber laser. A much more challenging task would be to obtain mode locking regime directly in master oscillators based on spinning ring gain elements. A detailed investigation of mode-locked performance of the spinning ring laser systems is a subject of future studies.

One obvious engineering difficulty needed to be solved in order to produce a practical laser system is the removal of heat from a moving gain element. Our current efforts in this area, where we already have very promising experimental results, are subject to our future publications.