Ultrafast laser sources operating in the mid-infrared (mid-IR) region, which contains the characteristic fingerprint spectra of many important molecules and transparent windows of atmosphere, are of significant importance in a variety of applications. Over the past decade, a significant progress has been made in the development of inexpensive, compact, high-efficiency mid-IR ultrafast mode-locked lasers in the picosecond and femtosecond domains that cover the 2.0 μm–3.5 μm spectral region. These achievements open new opportunities for applications in areas such as molecular spectroscopy, frequency metrology, material processing, and medical diagnostics and treatment. In this review, starting with the introduction of mid-IR mode-locking techniques, we mainly summarize and review the recent progress of mid-IR mode-locked laser sources, including Tm3+-, Ho3+-, and Tm3+/Ho3+-doped all-solid-state and fiber lasers for the 2.0 μm spectral region, Cr2+:ZnSe and Cr2+:ZnS lasers for the 2.4 μm region, and Er3+-, Ho3+/Pr3+-, and Dy3+-doped fluoride fiber lasers for the 2.8 μm–3.5 μm region. Then, some emerging and representative applications of mid-IR ultrafast mode-locked laser sources are presented and illustrated. Finally, outlooks and challenges for future development of ultrafast mid-IR laser sources are discussed and analyzed. The development of ultrafast mid-IR laser sources, together with the ongoing progress in related application technologies, will create new avenues of research and expand unexplored applications in scientific research, industry, and other fields.

Ultrafast mid-infrared (mid-IR) laser sources have recently attracted tremendous interest due to their promising applications in molecular spectroscopy, mid-IR supercontinuum generation, mid-IR frequency combs, material processing, laser surgery, biodiagnostics, etc.1–3 Passive mode locking is a commonly used, straightforward method to generate ultrashort laser pulses in various spectral regions. However, limited by the availability of gain materials and saturable absorbers (SAs) in the mid-IR region, passively mode-locked ultrafast mid-IR laser sources were seldom available before the early 2000s. In order to obtain mid-IR ultrashort pulses, the most common approaches were nonlinear conversion processes such as optical parametric oscillation,4 optical parametric amplification,5,6 and four-wave mixing.7–9 However, compared with the mode-locked lasers, especially the diode-pumped mode-locked lasers, these nonlinear approaches are usually very expensive and complex, and require precise optical synchronization, which prevents their widespread implementation. Thus, the development of high-efficiency, low-cost, compact mode-locked lasers with picosecond or femtosecond pulse durations, diffraction-limited beam quality, and high average power in the mid-IR region has been the goal pursued by laser researchers.

Over the past decade, benefiting from the rapid advances in material research, various mid-IR laser gain materials with excellent optical and mechanical properties have been developed with the aim to generate mid-IR ultrashort pulses.10–14 And significant developments have also been achieved in the mid-IR mode-locked laser sources, driven by new materials, technologies, and continued innovations in laser research field. Nowadays, less than 10-optical-cycle and even ∼3-optical-cycle ultrashort pulses have been directly generated in mode-locked laser oscillators around 2.0 μm and 2.4 μm regions, respectively,15,16 while the wavelength of the mode-locked lasers has been extended to ∼3.5 μm.17 These newly developed mid-IR ultrafast laser sources, together with ongoing progress in related technologies, can open new research avenues and offer novel opportunities for applications in scientific research, industry, and other fields. In this work, we review the progresses of mid-IR mode-locked lasers in the last decade, including all-solid-state and fiber mode-locked laser sources that operate around the 2.0 μm, 2.4 μm, and 2.8 μm–3.5 μm spectral regions.

The structure of this review is as follows. In Sec. II, the current main mid-IR mode-locking techniques are introduced. Section III presents and discusses the recent progress in mid-IR passively mode-locked lasers, including Tm3+-, Ho3+-, and Tm3+/Ho3+-doped mode-locked all-solid-state and fiber lasers around 2.0 μm, Cr2+:ZnSe and Cr2+:ZnS lasers around 2.4 μm, and recently developed Er3+-, Ho3+/Pr3+-, and Dy3+-doped fluoride fiber lasers in the 2.8 μm–3.5 μm region. In Sec. IV, some selected applications of ultrafast mid-IR lasers are presented to show the statuses and prospects in diverse application areas. Section V provides a summary and outlook regarding the development of mid-IR mode-locked laser sources.

Unlike active mode locking that typically produces longer (∼100 ps) pulses, passive mode locking has the ability to generate pulses as short as a few femtoseconds. In passive mode locking, a saturable absorption effect is typically required to modulate losses inside the laser cavity. Therefore, continuous-wave (CW) and low intensity laser radiation suffer non-negligible absorption losses, while short pulses with high peak power saturate the absorber and achieve low losses. Meanwhile, individual short pulses are shaped because the high-intensity portion at the pulse peak saturates the absorber more than its low-intensity wings, further shortening the pulse width in the time domain. With current technology, saturable absorption can be induced either by real SAs such as semiconductor saturable absorber mirrors (SESAMs) and various low-dimensional materials or via nonlinear optical effects such as the Kerr-lens effect, nonlinear polarization rotation (NPR), and nonlinear optical/amplifying loop mirrors (NOLM/NALM). These real SAs and nonlinear optical effects in passive mode-locked lasers tend to exhibit very fast (femtosecond to picosecond) response times, which aid in ultrashort pulse generation. Although they are similar to those used in near-infrared (near-IR) mode-locked lasers, passive mode-locking techniques for mid-IR lasers have several different characteristics and requirements, which are introduced in this section.

SAs are key to achieving picosecond and femtosecond pulse operation in both all-solid-state lasers and fiber lasers at various wavelengths. The SESAM was first demonstrated as a mode locker in an all-solid-state Nd3+:YLF laser in 1992.18 Since then, it has been widely used in various passively mode-locked lasers and is considered a revolutionary technology for ultrashort pulse generation. In particular, the broad design freedom associated with SESAMs provides compelling advantages such as flexible, precise parameter customization and robust functionality that help to satisfy specific laser requirements. SESAMs enable improvements in the output characteristics of various ultrafast lasers in different wavelength regions. However, due to lattice mismatch and growth-related issues,19 commercial GaAs-based SESAMs in the mid-IR region have not yet reached the same maturity level as those used in the near-IR region. Their relatively poor saturable absorption properties, especially relatively long absorption recovery times, inevitably degrade the laser mode-locking performance in the mid-IR region. In order to obtain SESAMs with better capabilities for mid-IR mode-locking, new GaSb-based SESAMs with appropriate postprocessing techniques, such as ion irradiation/implantation and incorporation of quantum wells or quantum dots, have been explored.20–22 Compared with standard GaAs-based devices, GaSb-based SESAMs can achieve better lattice-matching, exhibit lower nonsaturable losses, and offer significantly shorter recovery times. They have been used successfully in several mid-IR mode-locked all-solid-state and semiconductor disk lasers for picosecond and even sub-10 optical-cycle femtosecond pulse generation.15,19–30 Nevertheless, the limited operation wavelengths of GaSb-SESAMs hinder their applications in mid-IR mode-locked lasers beyond 3 μm.31,32

In addition to continued SESAM improvement and optimization, significant efforts have been made to exploit SAs based on novel materials. This has been done in order to develop reliable SAs with broader operating wavelengths (e.g., beyond 3.0 μm) and saturable absorption characteristics that are comparable to or better than those of current SESAMs. Newly emerged low-dimensional and bulk materials such as carbon nanotubes (CNTs),33–41 graphene,42–53 topological insulators (TIs),54–58 transition metal dichalcogenides (TMDCs),59 black phosphorus (BP),60–62 and topological Dirac semimetals63 have recently attracted significant attention and been extensively explored for mid-IR mode-locked lasers. Their advantages of broadband absorption region, ultrafast recovery time, low saturation fluence, and easy fabrication and integration make them potential candidates as SAs for various mid-IR lasers. However, some nonnegligible problems such as poor production repeatability and performance degradation in long-term applications remain to be solved.

1. Carbon nanotubes and graphene

Carbon-based low-dimensional materials, particularly CNTs and graphene, have shown significant promise as SAs that can generate ultrashort laser pulses at various wavelengths, including in the mid-IR region. Their unique electronic structures endow them with interesting linear and nonlinear optical properties,64–68 including broadband operating regions, large optical nonlinearities, ultrafast carrier relaxation times, and low saturation intensities. These properties make them attractive SESAM substitutes for ultrashort mid-IR pulse generation. CNTs are direct bandgap materials with excellent chemical stabilities and high optical damage thresholds. Their bandgaps depend on their diameters and chiralities.64 Natural mixtures of CNTs with various diameters and chiralities can provide a broadband operation region that covers approximately 0.8–2.4 μm and can be used as SAs for Tm3+-, Ho3+-, and Cr2+-doped mid-IR mode-locked lasers. Graphene, a two-dimensional allotrope of carbon atoms with a honeycomb lattice, has a zero bandgap and a linear energy dispersion in momentum space.69 These properties make it applicable as a SA for mode locking over an ultrawide spectral range from the visible to the mid-IR region. Compared with CNTs, graphene exhibits additional advantages of broader operating bandwidth, uniform linear and nonlinear optical response, and easy integration into photonic components and systems. Ultrafast pulse generation around the mid-IR regions of 2.0 μm,42 2.4 μm,47 and 2.8 μm53 has been demonstrated with graphene SAs. In addition, the fascinating electronic and optical properties of these carbon-based low-dimensional materials have created new opportunities for optoelectronic applications and triggered an explosion of research to explore other novel low-dimensional materials for mid-IR ultrashort pulse generation.

2. Topological insulators (TIs)

Over the past few years, TIs such as Bi2Te3, Bi2Se3, and Sb2Te3 have also emerged as promising broadband SA candidates for the near-IR to mid-IR regions.54–58 TIs represent a new class of electronic material in which the bulk state has a bandgap like that of an ordinary insulator but the surface or edge state is topologically protected conducting states and exhibits a Dirac cone-like electronic band structure.70,71 Due to a combination of spin–orbit interaction and time-reversal symmetry, TIs exhibit unique electronic and optical properties such as the quantum spin Hall effect,72 topological magnetoelectric effect,73 high nonlinear refractive indices,74 and ultrafast relaxation times.75 This new type of Dirac material was first successfully demonstrated as a SA in a passively mode-locked fiber laser around 1550 nm in 2012.76 Since then, several trials have been conducted by various groups to investigate their nonlinear absorption features in pulsed laser systems in the near-IR region. Due to Dirac-like linear dispersions which are similar to that of graphene, TIs can provide ultrabroadband operating bandwidth up to the mid-IR region. In addition, investigations of Bi2Se3, Bi2Te3, and Bi2SeTe2 indicate high modulation depths and relatively small nonsaturable absorption losses.74,77,78 With these unique properties, TIs are attractive SAs for mode-locked pulse generation in the mid-IR region.

3. Transition metal dichalcogenides (TMDCs)

TMDCs are semiconductors with MX2 stoichiometry, where M is a transition metal atom (e.g., Mo or W) and X is a chalcogen atom (e.g., S, Se, or Te). They have recently been demonstrated as SAs for mid-IR mode-locked lasers.79–85 TMDC bulk crystals are bonded together by weak van der Waals force and can be separated into monolayer and few-layer forms via mechanical exfoliation.86 Unlike graphene, TMDCs exhibit layer-dependent bandgaps that generally range from direct in monolayer to indirect in the bulk form.87 This creates new opportunities for photonic and optoelectronic applications in various wavelength regions.88–90 In addition, the optical properties of ultrafast carrier dynamics and high nonlinear optical response discovered in monolayer and few-layer TMDCs further support their development as SAs for mode-locked lasers. However, the optical bandgaps of most traditional 2H-phase TMDCs typically appear in the visible to near-IR spectral range due to interband optical transitions.88 They are generally not suitable for mid-IR photonic applications. Recently, Wang et al. demonstrated that wideband saturable absorption extending to the mid-IR region can be achieved in few-layer TMDCs by introducing suitable defects.91 This finding provides a feasible approach for the use of TMDCs in mid-IR mode-locked lasers. Meanwhile, some novel TMDC materials and new absorption mechanisms (e.g., sub-bandgap optical absorption) have been proposed and developed to scale the operating band into longer (3.0 μm to 10.0 μm) wavelength ranges.92–94 This significantly extends the application of these materials to mid-IR ultrafast laser technology.

4. Black phosphorus (BP)

Like TMDCs, BP is a layered material in which individual atomic layers are stacked via van der Waals interactions. It also demonstrates strong nonlinear absorption in various spectral regions. Monolayer BP is a semiconductor with a puckered honeycomb structure in which one phosphorus atom is covalently bonded to three adjacent phosphorus atoms.95 Few-layer black phosphorus has a direct bandgap,96 high carrier mobility (up to ∼1000 cm2 V−1 s−1),96–98 and a large on/off ratio (>105) at room temperature.96,99 These characteristics make it a promising two-dimensional semiconductor material for electronic and optoelectronic applications. In addition, theoretical and experimental investigations have shown that the direct bandgaps of the BP materials vary from ∼2.0 eV for a monolayer to ∼0.3 eV for bulk material due to layer-layer coupling,100,101 and this bandgap can be controlled by tuning the number of stacked layers. Such a layer-dependent direct bandgap can lead to a broad range of applications in photonics in the ∼0.6 μm to ∼4 μm region. Recently, broadband nonlinear optical absorption from 0.8 μm to 2.1 μm has been examined by Wang et al.,102 showing a clear wavelength-dependent nonlinear optical response. The combination of strong saturable absorption and a broadband operation bandwidth makes BP a promising candidate for ultrashort pulse generation across the visible, near-IR, and mid-IR spectral regions.

In addition to using real SAs, one can also exploit artificial SAs to generate saturable absorption effects. The most prominent artificial saturable absorption effect applicable to mode locking is the Kerr-lens effect, which is induced by Kerr nonlinearity.103 The physical origin of Kerr nonlinearity is nonlinear polarization generated in the medium under high electric field intensity. This phenomenon changes the refractive index inside the medium in response to the intensity of the applied electric field. Considering that the laser beam power densities of mode-locked lasers typically conform to Gaussian distributions, the refractive index change induced by the beam is greater in the center than around the edge, and thus produces a self-focusing effect referred to as the Kerr-lens effect. Since Kerr-lens mode locking was first demonstrated in a Ti:sapphire laser in 1991,104 it has become a common ultrashort-pulse generation technique in the near-IR spectral region. However, KLM is more challenging to achieve in the mid-IR wavelength range than in the near-IR because the magnitude of the Kerr-lens effect in the cavity is inversely proportional to the cavity mode area. Since the cavity mode radius is proportional to the square root of the laser wavelength, the Kerr-lens effect is not usually sufficient for KLM in mid-IR lasers, except for lasers whose gain materials have particularly high nonlinear refractive indices, such as Cr2+:ZnSe and Cr2+:ZnS. KLM remains an attractive mode locking technology for ultrashort few-cycle pulse generation in the mid-IR region because it offers extremely fast response time in the few-femtosecond range and ultrabroadband wavelength operation even though critical cavity alignment and initiation via external perturbations are usually required.

NPR is a commonly used artificial SA for mode-locked fiber lasers. Its saturable absorption effect arises from nonlinear phase delay-induced light polarization rotation. Linearly polarized light is converted to elliptically polarized light by a quarter waveplate before entering the fiber. This elliptical polarization can be treated as a combination of two orthogonal polarizations with unequal intensities. Self-phase modulation in the fiber introduces a certain amount of nonlinear phase delay between the two orthogonal polarizations, resulting in intensity-dependent polarization rotation within the fiber. Nonlinear amplitude modulation is achieved by rotating a half waveplate to select the high-intensity pulse peak through a polarizer while blocking low-intensity pulse edges and CW waves. NPR can not only initiate the mode-locking operation but also narrow the pulse for achieving shorter pulse generation. Since its first demonstration, NPR has been extensively applied to near-IR mode-locked fiber lasers.105,106 With the development of mid-IR fiber lasers, particularly fluoride fiber lasers that operate beyond 3 μm where commercial InGaAs-based SESAMs are unavailable, NPR has become increasingly essential due to its wavelength independence and simple structure. Currently, NPR is a common way of generating femtosecond mode-locked pulses in 3.0 μm fluoride fiber lasers where traditional SAs and pulse generation approaches are limited.

Another simple way to realize mode locking in fiber lasers is to implement a nonlinear optical loop mirror (NOLM), which provides intensity-dependent transmission via interference between two counter-propagating waves in a fiber loop. A laser beam is fed into one input port of an X fiber coupler and split into two waves that counterpropagate in the fiber loop, which return back to the coupler and interfere with each other. Transmission to the output port is determined by the interference condition, e.g., the phase delay between the two counterpropagating waves. If a coupler with unequal coupling is employed, the nonlinear phases induced by self-phase modulation are different for the two counterpropagating waves, resulting in a nonlinear phase delay. Since the amount of the nonlinear phase delay depends on the laser intensity, intensity-dependent transmission is formed and the system can act as an artificial fast SA for mode-locked lasers. By introducing an amplification fiber section at one end of the fiber loop in NOLM, a nonlinear amplifying loop mirror (NALM) is formed. It increases the difference between the accumulated nonlinear phase shifts of the two counterpropagating waves and, thus, significantly reduces the power required to initiate mode locking. Thus far, NOLM and NALM have been widely used in mode-locked fiber lasers that cover the 1.0 μm to 2.0 μm spectral region.107–111 With the development of fiber components and related technology, particularly the fluoride fiber-based X coupler and related splicing, there is potential to extend NOLM and NALM operating wavelengths to the longer mid-IR region.

The trivalent rare earth ions Tm3+ and Ho3+ are the most well-known elements that provide emissions for high-power CW and pulsed laser operation in the ∼2.0 μm wavelength range. Various types of high-efficiency and high-power laser emissions have been successfully demonstrated in a variety of host crystals, ceramics, and fibers around this “eye-safe” spectral region since the first Tm3+- and Ho3+-doped lasers were reported in the 1960s.112 In the pulsed operation regime, generation of mode-locked pulses at room temperature around 2.0 μm was first demonstrated in the Cr3+/Tm3+:YAG and Cr3+/Tm3+/Ho3+:YAG bulk lasers via an acousto-optic modulator in 1991, and 45 ps and 800 ps pulses were obtained, respectively.113 At almost the same time, 35 ps pulses were also achieved in an acousto-optic modulator mode-locked Tm3+:YAG laser with an average output power of 70 mW at 2.01 μm.114 A few years later, a milestone in the generation of ultrashort pulses around 2.0 μm was accomplished with a Tm3+-doped fiber ring laser that was passively mode-locked via additive-pulse mode locking,115 which produced 350–500 fs pulses at ∼1.8–1.9 μm. Since then, many efforts have been made to produce ultrashort pulses in the 2.0 μm spectral region using various lasers. In this section, recent progress in mode-locked all-solid-state and fiber lasers around 2.0 μm is reviewed.

1. Mode-locked all-solid-state lasers around 2.0 μm

Although mode-locked pulses with 190 fs pulse width were produced from a Tm3+-doped silica fiber laser as early as 1996,116 only active mode locking with tens to hundreds of picosecond pulses output was achieved in all-solid-state lasers in the following decade.117–121 For passive mode locking, which can generate ultrashort picosecond and femtosecond pulses, Z-folded or X-folded resonators were commonly used to achieve suitable laser mode sizes in the laser gain material and on the SA. However, due to the relatively low stimulated emission cross sections of the Tm3+- and Ho3+-doped bulk laser gain media, it is not a trivial task to suppress Q-switching instabilities and achieve stable CW mode-locking operation in 2.0 μm all-solid-state lasers. Another challenge to achieve mode locking is providing reliable SA, which is the key element and plays a crucial role in generating ultrashort mode-locked pulses.

Among various SAs mentioned in Sec. II, SESAM is the most commonly used one for passively mode-locked lasers. However, it was not mature enough to support stable mode locking in solid-state lasers in the 2.0 μm spectral region until the late 2000s due to the relatively poor performances in absorption recovery time, nonsaturable loss, and damage threshold. In 2009, an InGaAsSb-based quantum well SESAM was developed, in which GaSb/AlAsSb distributed Bragg reflectors and Te-doped GaSb substrates were adopted to avoid lattice mismatch, significantly improving the SESAM quality around the operating wavelength of 2.0 μm. With this optimized SESAM as the mode locker, Lagatsky et al. achieved passive mode locking in a Tm3+/Ho3+-doped KY(WO4)2 crystal laser.23 Short pulses with a duration of 3.3 ps and an average power of 315 mW were produced at 2057 nm. Subsequently, transform-limited 570 fs pulses were achieved with a lower output power of 123 mW in the same SESAM mode-locked laser by introducing a pair of fused silica prisms to compensate for intracavity dispersion.20 Since this first femtosecond operation of a mode-locked bulk laser was demonstrated, significant efforts have been made to generate ultrashort pulses with a shorter pulse duration toward less than 100 fs directly from mode-locked all-solid-state lasers around 2.0 μm. By using a Tm3+/Ho3+:NaY(WO4)2 disordered crystal which has broader and flatter emission spectra, 191 fs transform-limited mode-locked pulses around 2060 nm were produced with an 82 mW average output power.24 Later on, utilizing the smooth and broad gain of Tm3+-doped sesquioxide materials, Lagatsky et al. further reported soliton pulses as short as 105 fs centered at 2010 nm from a Tm3+:LuScO3 crystal laser in 2013. This represented considerable progress in the development of mode-locked all-solid-state lasers in the 2.0 μm spectral region.122 After that, passive mode locking was achieved in various Tm3+-, Ho3+-, and Tm3+/Ho3+-doped bulk laser materials with SESAM and semiconductor saturable absorbers as mode lockers. Although various mode-locked pulse durations from femtoseconds to dozens of picoseconds were achieved (Table I), no shorter mode-locked pulses were produced for nearly four years. Until 2018, the aim to break through sub-100 fs SESAM mode-locked pulses was successfully achieved by Wang et al. from a Tm3+:(Lu2/3Sc1/3)2O3 mixed ceramic laser.123 This milestone was achieved using a Ti:sapphire pump laser source and an InGaAsSb quantum well-based SESAM. Five chirped mirrors were implemented for precise intracavity dispersion management, as shown in Fig. 1. Sub-10-optical-cycle (63 fs) pulses were generated at 2.06 μm for the first time, which are the shortest pulses obtained from the all-solid-state lasers around 2.0 μm up until now. The obvious Kelly sidebands on the mode-locked spectrum shown in Fig. 1(c) demonstrate the soliton mode locking operation of the laser. In the soliton mode-locking regime, the obtained shortest pulses are mainly attributed to the balance between strong nonlinear self-phase modulation and net negative cavity dispersion, as well as the laser gain, losses, and gain bandwidth,124–126 rather than the SESAM which typically acts as a slow SA and is only used to start and stabilize the soliton pulses.

TABLE I.

Passively mode-locked all-solid-state lasers around the 2.0 μm spectral region. ML is the mode-locking, τp is the pulse width, λ0 is the center wavelength, Pout is the average output power, SESA is the semiconductor saturable absorber, TDFL is the Tm3+-doped fiber laser, EYDFA is the Er3+/Yb3+-doped fiber amplifier, NLM is the nonlinear mirror, and QD is the quantum dot.

Laser gain materialML techniqueτp (ps)λ0 (nm)Pout (mW)Pump sourceReferences
Tm3+:KLu(WO4)2 SWCNTs 9.7 1944 240 Ti:sapphire 33  
Tm3+/Ho3+:KY(WO4)2 SESAM 3.3 2057 315 Ti:sapphire 23  
 SESAM 0.570 2055 130 Ti:sapphire 20  
Tm3+:GdLiF4 SESA 17 1886 38 LD 128  
Cr3+/Tm3+/Ho3+:Y3Sc2Al3O12 PbS QD 290 2090 … Flashlamp 139  
Tm3+:GPNG glass SESAM 0.410 1997 62 Ti:sapphire 140  
Tm3+/Ho3+:TZN glass SESAM 0.630 2012 38 Ti:sapphire 140  
Tm3+:KLu(WO4)2 PbS QD … 1936 20 LD 141  
 SWCNTs 0.141 2037 26 Ti:sapphire 35  
Tm3+/Ho3+:NaY(WO4)2 SESAM 0.191 2060 82 Ti:sapphire 24  
Tm3+/Ho3+:YAG SESA 60 2091 160 Ti:sapphire 142  
 SESAM 21.3 2091 63 Ti:sapphire 19  
Ho3+:YLiF4 SESAM 1.1 2066 580 TDFL 143  
Tm3+:KY(WO4)2 SESAM 0.386 2029 235 Ti:sapphire 25  
Tm3+:Lu2O3 ceramic SESAM 0.180 2076 400 Ti:sapphire 26  
 SESAM 0.242 2068 500 LD 144  
 SESAM 0.300 2040 430 Ti:sapphire 122  
 Graphene 0.410 2067 270 Ti:sapphire 44  
Tm3+:Sc2O3 SESAM 0.218 2107 210 Ti:sapphire 29  
 KLM 0.166 2124 440 EYDFA 137  
Tm3+:CLNGG SESAM 0.479 1994 288 LD 129  
 Graphene 0.729 2018 60.2 LD 42  
 Graphene 0.882 2014.4 60 LD 43  
 Graphene 0.354 2010 97 LD 48  
 SWCNTs 0.078 2017 54 Ti:sapphire 39  
 MoS2 … 1979 62 LD 59  
Tm3+:Lu2O3 SWCNTs 0.175 2070 36 Ti:sapphire 34  
Tm3+:LuYSiO5 SESAM 19.6 1944.3 64.5 Ti:sapphire 145  
Tm3+/Ho3+:YVO4 SESAM 4.7 2041 151 LD 146  
Tm3+:LuScO3 SESAM 0.105 2010 … Ti:sapphire 122  
Tm3+/Ho3+:YAP SESAM 40.4 2064.5 132 Ti:sapphire 147  
Tm3+/Ho3+:KLu(WO4)2 SWCNTs 2.8 2059 97 Ti:sapphire 36  
 SESAM 4.2 2060 110 Ti:sapphire 21  
Tm3+:YAP NLM 4.7 1988 1050 LD 148  
 SESAM 1.89 1938 710 LD 149  
Tm3+:LuAG SESAM 38 2022.9 1210 LD 150  
 SESAM 13.6 2024 98 LD 31  
Tm3+:LuAG ceramic SESAM 2.7 2022 232 Ti:sapphire 151  
Tm3+:YAG ceramic SESAM 2.5 2012 40 Ti:sapphire 27  
 Graphene 2.8 2016 158 LD 46  
Tm3+:CaYAlO4 SESAM 35.3 1958.9 and 1960.6 830 LD 130  
 SESAM 0.621 1965 420 LD 132  
 SESAM 0.496 1975 531 LD 133  
 SESAM 53 1941.9 700 LD 152  
Tm3+:CaGdAlO4 SESAM 27 1949.5 330 LD 131  
 SESAM 0.646 2021 58 Ti:sapphire 28  
Tm3+/Ho3+:CaYAlO4 SWCNTs 0.096 2076.8 54 Ti:sapphire 40  
Ho3+:YAG KLM 0.220 2090 20 000 TDFL 136  
Ho3+:YAG ceramic SESAM 2.1 2064 10 TDFL 153  
 SESAM 241.5 2022.1 1840 TDFL 154  
Tm3+:MgWO4 Graphene 0.086 2017 39 Ti:sapphire 135  
Tm3+:YLF SESAM 94 2305.9 165 Ti:sapphire 155  
 KLM 0.514 2303 14.4 Ti:sapphire 138  
Tm3+:CNNGG SWCNTs 0.084 2017.7 22 Ti:sapphire 38  
Tm3+/Ho3+:CNGG SWCNTs 0.076 2081 67 Ti:sapphire 37  
Tm3+:(Lu2/3Sc1/3)2O3 ceramic SESAM 0.063 2057 34 Ti:sapphire 123  
Laser gain materialML techniqueτp (ps)λ0 (nm)Pout (mW)Pump sourceReferences
Tm3+:KLu(WO4)2 SWCNTs 9.7 1944 240 Ti:sapphire 33  
Tm3+/Ho3+:KY(WO4)2 SESAM 3.3 2057 315 Ti:sapphire 23  
 SESAM 0.570 2055 130 Ti:sapphire 20  
Tm3+:GdLiF4 SESA 17 1886 38 LD 128  
Cr3+/Tm3+/Ho3+:Y3Sc2Al3O12 PbS QD 290 2090 … Flashlamp 139  
Tm3+:GPNG glass SESAM 0.410 1997 62 Ti:sapphire 140  
Tm3+/Ho3+:TZN glass SESAM 0.630 2012 38 Ti:sapphire 140  
Tm3+:KLu(WO4)2 PbS QD … 1936 20 LD 141  
 SWCNTs 0.141 2037 26 Ti:sapphire 35  
Tm3+/Ho3+:NaY(WO4)2 SESAM 0.191 2060 82 Ti:sapphire 24  
Tm3+/Ho3+:YAG SESA 60 2091 160 Ti:sapphire 142  
 SESAM 21.3 2091 63 Ti:sapphire 19  
Ho3+:YLiF4 SESAM 1.1 2066 580 TDFL 143  
Tm3+:KY(WO4)2 SESAM 0.386 2029 235 Ti:sapphire 25  
Tm3+:Lu2O3 ceramic SESAM 0.180 2076 400 Ti:sapphire 26  
 SESAM 0.242 2068 500 LD 144  
 SESAM 0.300 2040 430 Ti:sapphire 122  
 Graphene 0.410 2067 270 Ti:sapphire 44  
Tm3+:Sc2O3 SESAM 0.218 2107 210 Ti:sapphire 29  
 KLM 0.166 2124 440 EYDFA 137  
Tm3+:CLNGG SESAM 0.479 1994 288 LD 129  
 Graphene 0.729 2018 60.2 LD 42  
 Graphene 0.882 2014.4 60 LD 43  
 Graphene 0.354 2010 97 LD 48  
 SWCNTs 0.078 2017 54 Ti:sapphire 39  
 MoS2 … 1979 62 LD 59  
Tm3+:Lu2O3 SWCNTs 0.175 2070 36 Ti:sapphire 34  
Tm3+:LuYSiO5 SESAM 19.6 1944.3 64.5 Ti:sapphire 145  
Tm3+/Ho3+:YVO4 SESAM 4.7 2041 151 LD 146  
Tm3+:LuScO3 SESAM 0.105 2010 … Ti:sapphire 122  
Tm3+/Ho3+:YAP SESAM 40.4 2064.5 132 Ti:sapphire 147  
Tm3+/Ho3+:KLu(WO4)2 SWCNTs 2.8 2059 97 Ti:sapphire 36  
 SESAM 4.2 2060 110 Ti:sapphire 21  
Tm3+:YAP NLM 4.7 1988 1050 LD 148  
 SESAM 1.89 1938 710 LD 149  
Tm3+:LuAG SESAM 38 2022.9 1210 LD 150  
 SESAM 13.6 2024 98 LD 31  
Tm3+:LuAG ceramic SESAM 2.7 2022 232 Ti:sapphire 151  
Tm3+:YAG ceramic SESAM 2.5 2012 40 Ti:sapphire 27  
 Graphene 2.8 2016 158 LD 46  
Tm3+:CaYAlO4 SESAM 35.3 1958.9 and 1960.6 830 LD 130  
 SESAM 0.621 1965 420 LD 132  
 SESAM 0.496 1975 531 LD 133  
 SESAM 53 1941.9 700 LD 152  
Tm3+:CaGdAlO4 SESAM 27 1949.5 330 LD 131  
 SESAM 0.646 2021 58 Ti:sapphire 28  
Tm3+/Ho3+:CaYAlO4 SWCNTs 0.096 2076.8 54 Ti:sapphire 40  
Ho3+:YAG KLM 0.220 2090 20 000 TDFL 136  
Ho3+:YAG ceramic SESAM 2.1 2064 10 TDFL 153  
 SESAM 241.5 2022.1 1840 TDFL 154  
Tm3+:MgWO4 Graphene 0.086 2017 39 Ti:sapphire 135  
Tm3+:YLF SESAM 94 2305.9 165 Ti:sapphire 155  
 KLM 0.514 2303 14.4 Ti:sapphire 138  
Tm3+:CNNGG SWCNTs 0.084 2017.7 22 Ti:sapphire 38  
Tm3+/Ho3+:CNGG SWCNTs 0.076 2081 67 Ti:sapphire 37  
Tm3+:(Lu2/3Sc1/3)2O3 ceramic SESAM 0.063 2057 34 Ti:sapphire 123  
FIG. 1.

(a) Schematic, (b) autocorrelation trace, and (c) optical spectrum of a mode-locked Tm3+:LuScO mixed ceramic laser. L: lens. M: dichroic mirror, CM: chirped mirror. OC: output coupler. r: radius of curvature. f: focal length. Reproduced with permission from Wang et al., Opt. Express 26, 10299 (2018). Copyright 2018 OSA Publishing.

FIG. 1.

(a) Schematic, (b) autocorrelation trace, and (c) optical spectrum of a mode-locked Tm3+:LuScO mixed ceramic laser. L: lens. M: dichroic mirror, CM: chirped mirror. OC: output coupler. r: radius of curvature. f: focal length. Reproduced with permission from Wang et al., Opt. Express 26, 10299 (2018). Copyright 2018 OSA Publishing.

Close modal

It is worthwhile to note that most of the 2.0 μm passively mode-locked all-solid-state lasers, particularly those with femtosecond operation, were pumped using Ti:sapphire lasers due to their good beam quality and high brightness. These high-brightness pump sources can well match a tightly focused laser mode. This helps to suppress a tendency toward Q-switched mode locking instability and aids in achieving stronger self-phase modulation inside the gain medium, benefitting for obtaining desired stable CW mode-locked operation and very short mode-locked pulses.127 However, the high costs and bulky volumes of these pump sources represent major obstacles to development of practical 2.0 μm mode-locked ultrafast lasers and significantly limit their extensive applications. With the rapid development of AlGaAs-based laser diodes (LDs) in recent years, direct pumping of Tm3+- and Tm3+/Ho3+-doped materials with compact, low-cost, efficient commercial LDs near 790 nm has become very attractive, even though it will be more challenging to suppress Q-switched mode locking and obtain shorter pulses than previous works pumped by the Ti:sapphire laser system. In 2010, LD-pumped passively mode-locked operation around 2.0 μm was achieved in a Tm3+:GdLiF4 laser with a transmission-type InGaAs quantum well SA. It generated 17 ps mode-locked pulses with an average power of only 38 mW.128 Two years later, Ma et al. presented the first diode-pumped femtosecond operation of a 2 μm mode-locked solid-state laser using a Tm3+:CLNGG disordered crystal as the gain medium.129 The laser generated stable 479 fs ultrashort mode-locked pulses with average output power as high as 288 mW at 1994 nm. The pump source was a 790 nm single-emitter AlGaAs LD, and mode locking was performed using a commercial SESAM. Since then, several LD-pumped mode-locking results with various laser gain materials and SAs in the 2.0 μm spectral region have been reported.42,43,46,59,130–133 Kong et al. reported the first dual-wavelength synchronously mode-locked laser around 2.0 μm in 2015, as shown in Fig. 2(a). Benefitting from the dual-wavelength emission of a Tm3+:CaYAlO4 disordered crystal, the mode-locked laser delivered temporally synchronous dual-color pulses at 1958.9 nm and 1960.6 nm [Fig. 2(b)].130 Optical beating between the dual-color pulses was clearly observable in the form of beating pulses with 3.5-ps duration and 0.13-THz repetition rate [Fig. 2(c)]. The dual-wavelength synchronously mode-locked laser provides a potential laser source for efficient generation of coherent terahertz radiation via optical difference frequency. Given that a Tm3+:CaYAlO4 disordered crystal has a broadband wavelength tuning range of 185 nm (from 1861 nm to 2046 nm),134 it is also considered a good candidate for femtosecond pulse generation around 2.0 μm. Using SESAM mode-locking, 496 fs pulses were achieved in a diode-pumped Tm3+:CaYAlO4 laser with an average output power of 531 mW at 1975 nm.133 Recently, Qiao et al. found that the diode-pumped femtosecond mode-locked Tm3+:CaYAlO4 laser could operate in the high-order transverse mode [Figs. 3(a)–3(c)].132 This is an interesting discovery because it is traditionally thought that femtosecond mode-locked lasers work only in the fundamental transverse mode. The high-order transverse-mode femtosecond laser provides a platform for generation of ultraclean, topological charge-tunable mid-IR femtosecond vortex beams [Figs. 3(d)–3(f)]. Figure 3(a) shows a schematic of a high-order transverse-mode femtosecond laser. Using a noncollinear pumping configuration, a pure and mode-order-tunable femtosecond Hermite-Gaussian beam was generated from the mode-locked laser. It was then converted into a femtosecond Laguerre-Gaussian vortex beam with a cylindrical lens mode converter, as shown in Fig. 3(d). The resulting femtosecond vortex beam had a donut profile and a clean intensity node with a ring-to-center intensity contrast of 36 dB [Fig. 3(f)], which approached the theoretical limit of a vortex beam. The ultraclean femtosecond vortex beam may create new possibilities for high-resolution nonlinear imaging, femtosecond trapping, and vortex chirped pulse amplification.

FIG. 2.

(a) Schematic, (b) optical spectrum, and (c) autocorrelation trace of a dual-wavelength mode-locked Tm3+:CaYAlO4 laser. Reproduced with permission from Kong et al., Opt. Lett. 40, 356 (2015). Copyright 2015 OSA Publishing.

FIG. 2.

(a) Schematic, (b) optical spectrum, and (c) autocorrelation trace of a dual-wavelength mode-locked Tm3+:CaYAlO4 laser. Reproduced with permission from Kong et al., Opt. Lett. 40, 356 (2015). Copyright 2015 OSA Publishing.

Close modal
FIG. 3.

Ultraclean femtosecond vortices from a tunable high-order transverse-mode femtosecond laser. (a) Schematic of the high-order transverse-mode femtosecond laser. LD: laser diode. L1 and L2: lenses. M1, M2, and M3: plano-concave highly reflective mirrors. OC: output coupler. (b) Autocorrelation trace and (c) optical spectrum of mode-locked pulses. (d) Schematic of Hermite Gaussian-to-Laguerre Gaussian mode converter and spatial intensity and phase interference setup. (e) Schematic of the spatial intensity contrast measurement apparatus. (f) Spatial intensity distribution of the femtosecond LG01 vortex in the far field. Reproduced with permission from Qiao et al., Opt. Lett. 42, 2547 (2017). Copyright 2017 OSA Publishing.

FIG. 3.

Ultraclean femtosecond vortices from a tunable high-order transverse-mode femtosecond laser. (a) Schematic of the high-order transverse-mode femtosecond laser. LD: laser diode. L1 and L2: lenses. M1, M2, and M3: plano-concave highly reflective mirrors. OC: output coupler. (b) Autocorrelation trace and (c) optical spectrum of mode-locked pulses. (d) Schematic of Hermite Gaussian-to-Laguerre Gaussian mode converter and spatial intensity and phase interference setup. (e) Schematic of the spatial intensity contrast measurement apparatus. (f) Spatial intensity distribution of the femtosecond LG01 vortex in the far field. Reproduced with permission from Qiao et al., Opt. Lett. 42, 2547 (2017). Copyright 2017 OSA Publishing.

Close modal

In recent years, low-dimensional materials have attracted tremendous research interest for use as next-generation optoelectronic devices such as photoelectric detectors, field-effect transistors, and optical modulators. As SAs for mid-IR mode-locked lasers, they generally have the advantages of broadband absorption, fast relaxation time, low saturation fluence, and simple fabrication methods. These characteristics make them potential replacements for SESAMs in ultrashort pulse generation. In fact, the first passively mode-locked operation of solid-state laser near 2.0 μm was achieved using a transmission-type single-walled carbon nanotube (SWCNT)-based SA in 2009.33 Stable mode-locked ∼10 ps pulses with a maximum average power of up to 240 mW were generated from a Tm3+:KLu(WO4)2 crystal pumped by a Ti:sapphire laser. After that, ultrashort mid-IR pulses with pulse durations from several picoseconds to less than 100 fs were achieved in various SWCNT-based mode-locked lasers,34–40 demonstrating the excellent saturable absorption performances of SWCNTs around the 2.0 μm region. At the same time, other low-dimensional material SAs such as graphene have also been successfully employed in all-solid-state mode-locked lasers to generate ultrashort mid-IR pulses. In 2012, Ma et al. presented the first graphene mode-locked femtosecond laser around 2.0 μm with a Tm3+:CLNGG disordered crystal,42 delivering stable 729 fs pulses at 2018 nm with an average output power of 60.2 mW. Later, graphene mode locking was also achieved in a Tm3+:Lu2O3 crystal laser,44 a Tm3+:LiYF4 crystal laser,45 and a Tm3+:YAG ceramic laser46 near 2.0 μm. Remarkably, with graphene SA, the first generation of sub-100 fs pulses from solid-state oscillators around 2.0 μm was achieved in a Tm3+:MgWO4 laser by a research group from Max-Born-Institute in 2017.135 Near transform-limited pulses as short as 86 fs operated at 2017 nm were produced by using chirped mirrors for dispersion compensation, as shown in Fig. 4. Since then, these researchers have demonstrated a series of ultrashort sub-100 fs mode-locked lasers in various Tm3+- and Tm3+/Ho3+-doped materials with SWCNTs37–40 and SESAMs.123 They have even propelled the pulse duration down to sub-10 optical cycles.123 The laser materials used in these works generally had flat, smooth, broadband gain spectra profiles that could support ultrashort femtosecond pulse generation. In addition, the tightly focused laser mode matched with the high-brightness Ti:sapphire pumping beam in the laser media, allowing for suppression of Q-switched mode locking and achieving strong self-phase modulation during laser operation. Moreover, all of these ultrashort sub-100 fs pulses were obtained with very low output coupling, resulting in high pulse energy in the cavities. Given the high energy and ultrashort duration of the pulse in the cavity, strong self-phase modulation could be induced to further broaden the mode-locked pulse spectrum. With the balance between self-phase modulation and anomalous group delay dispersion, ultrashort soliton mode-locked pulses were finally achieved. It should be noted that the pulse shaping was dominated by the interplay between group delay dispersion and self-phase modulation under these conditions, and the SA was no longer the crucial factor that determined the pulse duration.

FIG. 4.

(a) Measured second harmonic generation FROG traces. (b) Temporal intensity and phase. Black curve: Fourier-limited pulse profile. (c) Retrieved second harmonic generation FROG traces. (d) Spectral intensity and phase. Black curve: Directly measured spectrum by an InGaAs spectrometer. (e) Measured intensity autocorrelation curve fitted by a Sech2-shaped pulse and the autocorrelation curve of pulses retrieved from FROG trace. Inset: Autocorrelation trace in a 50-ps time window. Reproduced with permission from Wang et al., Opt. Lett. 42, 3076 (2017). Copyright 2017 OSA Publishing.

FIG. 4.

(a) Measured second harmonic generation FROG traces. (b) Temporal intensity and phase. Black curve: Fourier-limited pulse profile. (c) Retrieved second harmonic generation FROG traces. (d) Spectral intensity and phase. Black curve: Directly measured spectrum by an InGaAs spectrometer. (e) Measured intensity autocorrelation curve fitted by a Sech2-shaped pulse and the autocorrelation curve of pulses retrieved from FROG trace. Inset: Autocorrelation trace in a 50-ps time window. Reproduced with permission from Wang et al., Opt. Lett. 42, 3076 (2017). Copyright 2017 OSA Publishing.

Close modal

Compared to mode-locking techniques based on real SAs, KLM shows a significant advantage in generating shorter pulses due to its intensity response rather than energy fluence response. To obtain ultrashort pulses around 2.0 μm, KLM technique has also been tried in various Tm3+-, Ho3+-, and Tm3+/Ho3+-doped laser materials. However, due to the larger divergence of the mid-IR beam, Kerr-lens effect is usually not strong enough to achieve KLM in mid-IR lasers, except for the lasers whose gain materials have remarkably high nonlinear refractive indices, such as ZnSe and ZnS. Thus, it is generally a challenging work to achieve KLM in the 2.0 μm spectral region. Until 2017, remarkable progresses were obtained. Zhang et al. introduced a 1 mm-thick sapphire plate as the Kerr medium to provide the necessary Kerr self-focusing effect in a Ho3+:YAG thin-disk oscillator, and successfully realized KLM, delivering 220 fs pulses at 2090 nm with a record average output power of 20 W.136 Later, benefitting from the high nonlinear refractive index of Re2O3 material, KLM operation was also demonstrated in a Tm3+:Sc2O3 crystal laser, which was in-band pumped by an Er3+/Yb3+-doped fiber master oscillator power amplifier.137 Pulses as short as 166 fs with an average output power of 440 mW were obtained at 2124 nm. At almost the same time, Canbaz et al. used KLM to produce a new source of 2.3 μm femtosecond pulses from a Tm3+:YLF laser.138 An undoped 2 mm-thick polycrystalline ZnSe plate was used in the cavity to enhance the Kerr effect and help to achieve KLM. The mode-locked laser generated 514 fs pulses with an average power of 14.4 mW at 2.3 μm, which is believed to be the longest mode-locking wavelength from Tm3+-doped solid-state lasers to date.

Table I summarizes the main results of passively mode-locked Tm3+-, Ho3+-, and Tm3+/Ho3+-doped all-solid-state lasers published over the past decade. For each reference, we quote the shortest pulse width result.

2. Mode-locked fiber lasers around 2.0 μm

Compared to solid-state laser materials, optical fiber is an excellent platform that is capable of generating laser emission with many exclusive advantages such as compactness, high power, excellent beam quality, and outstanding heat dissipation capability. Moreover, the guiding property of fiber systems allows one to easily integrate fiber components (e.g., fiber gratings and fiber couplers) to build all-fiber laser systems. These systems are compact, robust, economical, and require minimal maintenance. With these advantages, fiber-based laser systems have attracted significant attention and been well developed in the near-IR wavelength region in recent decades. Their output power has been scaled up to kilowatt-class and their pulse widths span the range from nanoseconds to femtoseconds. In the mid-IR spectral region, ultrashort pulse generation in fiber lasers began in the 1990s when Nelson et al. demonstrated a Tm3+-doped mode-locked fiber laser that produced sub-500 fs pulses around 2.0 μm.115 Since then, ultrafast fiber lasers have evolved rapidly toward ever-higher performances in this spectral region, and a number of mode-locked Tm3+-, Ho3+-, and Tm3+/Ho3+-doped fiber laser systems have been reported. Since a few reviews have been published on fiber laser sources around 2.0 μm,156–159 our discussion focuses primarily on critical results and recent progress in mode-locked fiber lasers in this wavelength region.

In 1995, the first 2.0 μm femtosecond fiber laser was achieved using the NPR mode-locking technique. It produced 350 fs–500 fs pulses over a tuning range of 1.8 μm–1.9 μm.115 Subsequently, Sharp et al. presented an InGaAs SESAM mode-locked Tm3+-doped silica fiber laser capable of generating 190 fs pulses with an emission wavelength extending to 1.997 μm.116 A specific feature of optical fiber operating around 2.0 μm is that silica fibers exhibit anomalous dispersion, which causes typical soliton operation without dispersion management. The generated ultrashort soliton pulses were the results of balance between group delay dispersion and self-phase modulation, and no critical SA parameters were required under these conditions. However, due to the lack of optical fiber components in the 2.0 μm range, there were few mode-locked results reported in the following decade. Until 2007, transform-limited 650 fs–850 fs soliton pulses were achieved in a Tm3+/Ho3+-doped fiber oscillator with a Sb-based SESAM.160 Since then, SESAM mode locking has been successfully demonstrated in a series of fiber lasers near 2.0 μm,161–168 including the first Tm3+/Ho3+-doped fiber oscillator operating beyond 2.0 μm,163 the first diode-pumped mode-locked Ho3+-doped fiber laser,167 and a mode-locked Tm3+-doped multimode fiber oscillator with a record average power that exceeded 10 W.168 In the meantime, the booming development of low-dimensional nanomaterials over the past decade has triggered an explosion of investigation into their saturable absorption properties as SAs for mid-IR mode-locked fiber lasers. CNTs,169–174 graphene,175–184 TIs,185–188 TMDCs,79–83,189 BP,190–192 and other novel materials193–195 have been used to achieve mode-locking operation in various fiber lasers around 2.0 μm. They have generated various pulse patterns such as conservative solitons, dispersion-managed solitons, dissipative solitons, bound-state solitons, vector solitons, and harmonic soliton pulses. Among them, with a normally dispersive germanium-silicate fiber to control intracavity dispersion, 152 fs ultrashort pulses with a 52.8 nm bandwidth were generated in a Tm3+-doped all-fiber oscillator mode locked by CNTs.173 This was the shortest pulse duration with the broadest bandwidth directly obtained from CNT mode-locked 2.0 μm all-fiber lasers. Implementing dispersion management in this laser yielded obvious improvement in pulse duration. Stretched-pulse operation, which could overcome the pulse energy limitation of the fundamental soliton regime, was also achieved in a graphene mode-locked Ho3+-doped all-fiber laser.184 It delivered a high pulse energy reaching to 2.55 nJ with 190 fs pulse duration at 2060 nm. In addition, ultralow noise 58 fs soliton pulses were produced196 using a graphene electro-optic modulator and a real SA. This is the shortest pulse obtained from Tm3+-doped silica fiber oscillators thus far.

Another important mode-locking technology for ultrafast fiber lasers is use of an artificial SA based on nonlinear phase shift, such as NPR,115,197–202 NOLM,111 and NALM.110,164 Compared to real SAs, these mode-locking approaches have the advantages of ultrafast response time and wavelength-insensitive characteristic. In particular, NPR has been demonstrated to be a powerful technique that is able to generate versatile ultrashort pulses in mode-locked fiber lasers around 2.0 μm although bulk components are usually required in the laser setup. To overcome the soliton energy limitation in the NPR mode-locked fiber laser demonstrated by Nelson et al.,115 Engelbrecht et al. introduced a pair of gratings within the cavity to compensate for anomalous dispersion of the fiber, achieving a 4.3 nJ pulse energy.203 The generated 1.2 ps chirped pulses were centered at 1976 nm and could be externally compressed to a minimum pulse duration of 294 fs. By further shifting the cavity dispersion to the normal regime, dissipative soliton operation and self-similar pulse evolution could be obtained. These conditions allow the best combinations of high energy and short pulse duration. By employing ultrahigh numerical aperture fibers with normal dispersions, Tang et al. presented a Tm3+-doped fiber laser operating at a large normal dispersion with hybrid self-similar evolution and achieved 7.6 nJ, 130 fs pulses.200 Pulses as short as 160 fs with pulse energy above 1 nJ at 2060 nm were also produced in a Ho3+-doped fiber oscillator operating in the dispersion-managed soliton regime. Ultrashort sub-100 fs pulses were further obtained after passing through an additional nonlinear compressor stage.204 In 2014, an ultrafast mode-locked Tm3+-doped ZBLAN fiber laser oscillator was produced via NPR (Fig. 5), producing 45 fs pulses via external compression with an average power of 13 mW.199 This represents the shortest pulses generated from fiber laser oscillators operating around the 2 μm wavelength region up until now.

FIG. 5.

(a) Schematic of the experimental setup. TDF: Tm3+-doped fiber; SMF: Single-mode fiber; DM: Dichroic mirror; PBS: Polarizing beam splitter; HWP: Half-wave plate; QWP: Quarter-wave plate; GR: Grating. (b) Measured FROG trace. (c) FROG trace retrieved from the experimental FROG trace. (d) Retrieved pulse shape. (e) Retrieved spectral profile (filled blue curve) and phase (dashed red curve). Reproduced with permission from Y. Nomura and T. Fuji, Opt. Express 22, 12461 (2014). Copyright 2014 OSA Publishing.

FIG. 5.

(a) Schematic of the experimental setup. TDF: Tm3+-doped fiber; SMF: Single-mode fiber; DM: Dichroic mirror; PBS: Polarizing beam splitter; HWP: Half-wave plate; QWP: Quarter-wave plate; GR: Grating. (b) Measured FROG trace. (c) FROG trace retrieved from the experimental FROG trace. (d) Retrieved pulse shape. (e) Retrieved spectral profile (filled blue curve) and phase (dashed red curve). Reproduced with permission from Y. Nomura and T. Fuji, Opt. Express 22, 12461 (2014). Copyright 2014 OSA Publishing.

Close modal

Table II summarizes the important results of passively mode-locked Tm3+-, Ho3+-, and Tm3+/Ho3+-doped fiber lasers. For each reference, we quote the shortest pulse width result.

TABLE II.

Passively mode-locked Tm3+-, Ho3+-, and Tm3+/Ho3+-doped fiber lasers around the 2.0 μm spectral region. ML is the mode-locking, τp is the pulse width, λ0 is the center wavelength, Pout is the average output power, EDFL is an Er3+-doped fiber laser, EDFA is an Er3+-doped fiber amplifier, EYDFL is an Er3+/Yb3+-doped fiber laser, EYDFA is an Er3+/Yb3+-doped fiber amplifier, and TDFL is a Tm3+-doped fiber laser.

Laser gain materialML techniqueτp (fs)λ0 (nm)Pout (mW)Pump sourceReferences
Tm3+ silica fiber CNTs 152 1927 4.9 1570 nm EDFL 173  
 CNTs 1320 1932 3.4 1570 nm EDFL 169  
 Graphene 370 1902 1.7 1552 nm EYDFL 205  
 Graphene 3600 1940 2.6 … 175  
 Graphene Oxide … 2007 1.8 790 nm LD 206  
 Sb2Te3 890 1945 1.2 EDFA 188  
 WTe2 1250 1916 39.9 1550 nm LD 80  
 MoTe2 952 1930 36.7 1550 nm LD 81  
 WSe2 1160 1864 32.5 1550 nm EDFL 82  
 MoS2 1510 1926 1570 nm 85  
 BP 793 1910 1.5 1568 nm EYDFA 190  
 SESAM 190 1897 0.4 786 nm LD 116  
 SESAM 350 ∼1980 10 798 nm LD 164  
 SESAM 579 1918 158 1550 nm EYDFA 165  
 SESAM 1500 1980 10 798 nm LD 161  
 SESAM 15 800 1950 185 793 nm LD 207  
 NPR 350 1825 0.24 Ti:sapphire 115  
 NPR 173 1974 165.6 793 nm LD 197  
 NPR 119 1912 7.8 1575 nm EDFL 198  
 NPR 130 1925 203 1569 nm EDFL 200  
 NPR 330 1950 36.3 1560 nm EDFA 201  
 NPR 329 1990 35 1560 nm EDFA 202  
 … 58 ∼1925 506 1560 nm EDFA 196  
 NALM 680 2040 41 786 nm LD 164  
 NALM 1500 2034 0.7 786 nm LD 110  
 NOLM 2800 2017 0.1 793 nm LD 111  
Tm3+/Ho3+ silica fiber CNTs 2380 1860 1560 nm EDFA 174  
 Bi2Te3 795 1935 1.0 1550 nm LD 185  
 WS2 1300 1941 0.6 1550 nm LD 189  
 MoSe2 920 1912 4.3 1550 nm LD 79  
 SESAM 315 1968 22 Ti:sapphire 166  
 SESAM 1100 2060 10 799 nm LD 163  
Ho3+ silica fiber Graphene 190 2060 54 1950 nm TDFL 184  
 SESAM 2230 2094 28 1150 nm LD 167  
 NPR 160 2060 40 1950 nm TDFL 204  
Tm3+ fluoride fiber NPR 45 1890 13 Ti:sapphire 199  
Laser gain materialML techniqueτp (fs)λ0 (nm)Pout (mW)Pump sourceReferences
Tm3+ silica fiber CNTs 152 1927 4.9 1570 nm EDFL 173  
 CNTs 1320 1932 3.4 1570 nm EDFL 169  
 Graphene 370 1902 1.7 1552 nm EYDFL 205  
 Graphene 3600 1940 2.6 … 175  
 Graphene Oxide … 2007 1.8 790 nm LD 206  
 Sb2Te3 890 1945 1.2 EDFA 188  
 WTe2 1250 1916 39.9 1550 nm LD 80  
 MoTe2 952 1930 36.7 1550 nm LD 81  
 WSe2 1160 1864 32.5 1550 nm EDFL 82  
 MoS2 1510 1926 1570 nm 85  
 BP 793 1910 1.5 1568 nm EYDFA 190  
 SESAM 190 1897 0.4 786 nm LD 116  
 SESAM 350 ∼1980 10 798 nm LD 164  
 SESAM 579 1918 158 1550 nm EYDFA 165  
 SESAM 1500 1980 10 798 nm LD 161  
 SESAM 15 800 1950 185 793 nm LD 207  
 NPR 350 1825 0.24 Ti:sapphire 115  
 NPR 173 1974 165.6 793 nm LD 197  
 NPR 119 1912 7.8 1575 nm EDFL 198  
 NPR 130 1925 203 1569 nm EDFL 200  
 NPR 330 1950 36.3 1560 nm EDFA 201  
 NPR 329 1990 35 1560 nm EDFA 202  
 … 58 ∼1925 506 1560 nm EDFA 196  
 NALM 680 2040 41 786 nm LD 164  
 NALM 1500 2034 0.7 786 nm LD 110  
 NOLM 2800 2017 0.1 793 nm LD 111  
Tm3+/Ho3+ silica fiber CNTs 2380 1860 1560 nm EDFA 174  
 Bi2Te3 795 1935 1.0 1550 nm LD 185  
 WS2 1300 1941 0.6 1550 nm LD 189  
 MoSe2 920 1912 4.3 1550 nm LD 79  
 SESAM 315 1968 22 Ti:sapphire 166  
 SESAM 1100 2060 10 799 nm LD 163  
Ho3+ silica fiber Graphene 190 2060 54 1950 nm TDFL 184  
 SESAM 2230 2094 28 1150 nm LD 167  
 NPR 160 2060 40 1950 nm TDFL 204  
Tm3+ fluoride fiber NPR 45 1890 13 Ti:sapphire 199  

In the above section, we have discussed the ultrafast mode-locked laser sources around the 2.0 μm spectral region. Simultaneously, mid-IR mode-locked lasers operating around 2.4 μm have also attracted significant interest over the past decade due to their wide potential applications such as spectroscopy, gas sensing, high harmonic generation, etc. In this longer mid-IR wavelength region, transition-metal Cr2+-doped ZnSe and ZnS, which were introduced in the 1990s by Lawrence Livermore National Laboratory,208 represent a simple and viable route to achieve high-efficiency, high-power, cost-effective ultrafast mid-IR laser sources.

Cr2+:ZnSe and Cr2+:ZnS have highly similar spectroscopic properties. They feature four-level energy structures, broad vibronic emission spectra, the absence of excited state absorption, close to 100% fluorescence quantum efficiency at room temperature, good chemical and mechanical stability, and high thermal conductivity. These advantages, especially the ultrabroad emission spectrum, make them dominant media for femtosecond pulse generation around the 2 μm–3 μm spectral region and be referred to as “Ti:sapphire of the mid-IR.” Thus far, ultrabroad tuning ranges of 1973 nm–3349 nm in Cr2+:ZnSe and 1962 nm–3195 nm in Cr2+:ZnS have been obtained,11,209–212 which means they are very good candidates for ultrashort femtosecond mode-locked pulse generation. In addition, due to the broad absorption band around the 1.5 μm–2.0 μm wavelength region, they can be conveniently pumped by cost-effective and reliable Er3+- and Tm3+-doped fiber lasers.

Since the first Cr2+:ZnSe laser was demonstrated in 1996,208 significant progress has been made with regard to various Cr2+:ZnSe and Cr2+:ZnS laser systems, covering a broad range of operation modes and output parameters. The ultrabroad tuning operating regime is particularly attractive because the broad tuning range is not only useful for mid-IR spectroscopic and sensor applications but also very important for the gain medium to generate few-cycle ultrashort femtosecond pulses with passive mode locking techniques. In 2005, a 1100 nm tuning bandwidth from 2000 nm to 3100 nm was demonstrated in a CW Cr2+:ZnSe laser.213 To achieve broader wavelength tuning range, an intracavity-pumping configuration was used in a pulsed Cr2+:ZnSe laser, and 1220 nm smooth and continuous tuning from 1880 nm to 3100 nm was reported.210 After that, Sorokin et al. demonstrated the broadest tunable operation of Cr2+:ZnSe and Cr2+:ZnS materials with over 1400 nm wavelength tuning range in 2010.211 These extremely broad tuning ranges indicate the potential of Cr2+:ZnSe and Cr2+:ZnS to generate few- to single-cycle pulses in the mid-IR region around 2.4 μm.

Pumped by a CW NaCl:OH color center laser at 1.58 μm, the first mode-locked Cr2+:ZnSe laser was demonstrated with an acousto-optic modulator in 2000. It generated transform-limited Gaussian-shaped 4.4 ps mode-locked pulses at 2470 nm.214,215 Next, active mode locking was achieved in Cr2+:ZnSe single crystal216 and ceramic217 using an acousto-optic modulator. The mode-locked pulses obtained around 2.5 μm were approximately 4 ps and 21 ps in duration, respectively. In comparison to active mode locking, passive mode locking is the mainstream of ultrashort-pulse generation technique and can produce shorter pulses in various spectral regions. In 2005, the first passively mode-locked Cr2+:ZnSe laser was produced by Pollock et al. using a SESAM as the mode locker. It generated 10.8 ps mode-locked pulses near 2.5 μm.218 Since then, SESAM mode-locked Cr2+:ZnSe and Cr2+:ZnS ceramic and single crystal lasers have been continuously reported, delivering mode-locked pulses with various pulse widths in the 2.4 μm region.219–224 Shortly after the first SESAM mode-locked Cr2+:ZnSe laser was demonstrated, Sorokina et al. presented a passively mode-locked Cr2+:ZnS laser with an InAs/GaSb-based multiple quantum well SESAM. It produced ∼1.1 ps pulses in the 2.45 μm region.219 Subsequently, ultrashort 106 fs pulses were obtained by introducing a 5 mm sapphire plate for intracavity dispersion compensation under similar cavity conditions.220 In 2007, the same group used two chirped mirrors and a 3.1 mm YAG plate to further compensate for intracavity dispersion and achieved shorter pulses in a SESAM mode-locked Cr2+:ZnSe laser.221 Ultrashort ten-optical-cycle 80 fs pulses around 2.4 μm were obtained, which are the shortest pulses achieved from the SESAM mode-locked Cr2+:ZnSe and Cr2+:ZnS lasers thus far.

Compared to the SESAM mode-locking technique, KLM is more suitable to generation of ultrashort few-cycle mode-locked pulses due to its ultrafast response time and ultrabroad operation bandwidth. It is especially applicable to Cr2+:ZnSe and Cr2+:ZnS materials since they have high nonlinear refractive indices and particularly broadband emission spectra. KLM operation around 2.4 μm was first achieved in Cr2+:ZnSe lasers by two independent groups in 2009.225,226 They achieved 95 fs and 100 fs mode-locked soliton pulses in the experiments, respectively. Later, by replacing the MaF2 prism pair with a CaF2 prism pair for dispersion compensation, the mode-locked Cr2+:ZnSe laser delivered shorter 92 fs pulses.227 In 2013, with Cr2+:ZnS gain medium and KLM technique, transform-limited pulses with 69 fs pulse width and 550 mW average power were produced at 2.39 μm.228 Subsequently, the average power was increased to 1 W in a shorter cavity, generating mode-locked pulses with 75 fs pulse duration.229,230 In recent years, polycrystalline Cr2+:ZnSe and Cr2+:ZnS have attracted significant interest due to their advantages of high dopant concentrations, uniform dopant distributions, low losses, and mass-production with preassigned parameters. These properties support high-efficiency, high-power laser operation. In addition, because of their disordered polycrystalline structures that consist of a multitude of microscopic single-crystal grains with a broad distribution of grain sizes and orientations, a random quasi-phase-matching process can be achieved. This random quasi-phase-matching process has been demonstrated in several polycrystalline Cr2+:ZnSe and Cr2+:ZnS KLM lasers and enables direct nonlinear frequency conversion within the laser gain medium.16,231,232 Since the first KLM polycrystalline Cr2+:ZnSe and Cr2+:ZnS lasers were reported in 2014,233 the output parameters of ultrafast KLM polycrystalline Cr2+:ZnS and Cr2+:ZnSe lasers have been significantly improved.16,231–235 In 2015, ultrashort 29 fs mode-locked pulses were obtained from a KLM polycrystalline Cr2+:ZnS laser (Fig. 6),16 which are believed to be the shortest pulses obtained directly from mode-locked lasers in the 2.4 μm region. Under 4.6 W of pump power, the mode-locked laser had a spectrum as broad as 950 nm (1900 nm–2850 nm) at −30 dB level, which is a good source for mid-IR spectroscopy.

FIG. 6.

(a) Detected pulse train from the KLM Cr2+:ZnS laser at 100 MHz. (b) Net group delay dispersion of the resonator at 100 MHz. Inset: Beam profile of the mode-locked laser. (c) Measured spectrum of pulses and net reflectivity of the resonator's optics at 100 MHz. The gray background presents the transmission of standard air in a 1.5 m-long resonator. Emission spectra of optimized KLM lasers presented in logarithmic scale at (d) 300 MHz and (e) 100 MHz repetition rates: (i) fundamental mid-IR band, (ii) second harmonic generation band, (iii) third optical harmonic, (iv) fourth optical harmonic, (v) sum frequency generation between femtosecond mid-IR pulses and CW pump radiation, (vi) residual pumping at 1567 nm. Autocorrelations of optimized KLM laser at 100 MHz repetition rate: (f) measured without compensation, (g) measured with compensation. Reproduced with permission from Vasilyev et al., Opt. Lett. 40, 5054 (2015). Copyright 2015 OSA Publishing.

FIG. 6.

(a) Detected pulse train from the KLM Cr2+:ZnS laser at 100 MHz. (b) Net group delay dispersion of the resonator at 100 MHz. Inset: Beam profile of the mode-locked laser. (c) Measured spectrum of pulses and net reflectivity of the resonator's optics at 100 MHz. The gray background presents the transmission of standard air in a 1.5 m-long resonator. Emission spectra of optimized KLM lasers presented in logarithmic scale at (d) 300 MHz and (e) 100 MHz repetition rates: (i) fundamental mid-IR band, (ii) second harmonic generation band, (iii) third optical harmonic, (iv) fourth optical harmonic, (v) sum frequency generation between femtosecond mid-IR pulses and CW pump radiation, (vi) residual pumping at 1567 nm. Autocorrelations of optimized KLM laser at 100 MHz repetition rate: (f) measured without compensation, (g) measured with compensation. Reproduced with permission from Vasilyev et al., Opt. Lett. 40, 5054 (2015). Copyright 2015 OSA Publishing.

Close modal

Low-dimensional materials, such as graphene and CNTs, have also been demonstrated as promising SAs for Cr2+:ZnSe and Cr2+:ZnS mode-locked lasers around 2.4 μm in recent years. Using monolayer graphene deposited on a CaF2 plate as a SA, graphene mode-locking was first achieved in a Cr2+:ZnSe laser around 2500 nm in 2013.47 Mode-locked 226 fs pulses with an average output power of 80 mW were produced. Given that graphene material had ultrabroadband saturable absorption, Ma et al. fabricated a graphene-gold film saturable absorber mirror (GG-SAM) with a demonstrated operating bandwidth that exceeded 1300 nm and achieved mode locking in various all-solid-state lasers.48Figures 7(a)–7(c) show the structure, image, and characteristics of the GG-SAM. Based on the GG-SAM, a tunable femtosecond Cr2+:ZnSe laser was realized. It produced ultrashort pulses of 116 fs and supported a mode-locking tuning range of 116 nm (2310 nm–2426 nm), as shown in Figs. 7(d)–7(f). In 2014, mode-locking operation in the positive dispersion regime around 2.4 μm was achieved in a graphene mode-locked Cr2+:ZnS laser, which produced chirped pulses with a high pulse energy of 15.5 nJ.49 After compression by a pair of low OH-content infrasil prisms, 189 fs pulses were obtained. Later, the combination of a YAG wedge pair and a chirped mirror was used for dispersion compensation, and ultrashort 41 fs pulses centered at 2.4 μm with 190 nm spectral bandwidth were obtained in a similar graphene mode-locked Cr2+:ZnS laser.50 To improve the average output power, a ceramic Cr2+:ZnS material was employed as the gain medium. The resulting 140 fs mode-locked pulses around 2.35 μm had an average power as high as 1.0 W.51 Besides graphene, CNTs which generally operate around 1.0 μm–2.0 μm have also been shown to support mode-locked operation in the 2.4 μm spectral region. In 2014, a Cr2+:ZnS laser mode-locked with a CNT SA was reported for the first time. The resulting laser pulses had a high average power of 950 mW and a pulse duration as short as 61 fs at 2.35 μm,41 indicating the excellent performance of a CNT SA around 2.4 μm. The obtained result is the shortest pulse produced from CNTs mode-locked lasers so far.

FIG. 7.

(a) Structure of the GG-SAM. (b) The reflectivity spectra of the GG-SAM and gold-film substrate, and transmission spectrum of monolayer graphene on a quartz substrate. Inset: Photograph of the GG-SAM. (c) Raman spectrum of the GG-SAM after subtraction of the contribution of the gold film substrate. (d) Autocorrelation trace and (e) optical spectrum of the mode-locked Cr2+:ZnSe laser. (f) Tunable spectra of the mode-locked Cr2+:ZnSe laser. Reprinted with permission from Ma et al., Sci. Rep. 4, 5016 (2014). Copyright 2014 Nature Publishing Group.

FIG. 7.

(a) Structure of the GG-SAM. (b) The reflectivity spectra of the GG-SAM and gold-film substrate, and transmission spectrum of monolayer graphene on a quartz substrate. Inset: Photograph of the GG-SAM. (c) Raman spectrum of the GG-SAM after subtraction of the contribution of the gold film substrate. (d) Autocorrelation trace and (e) optical spectrum of the mode-locked Cr2+:ZnSe laser. (f) Tunable spectra of the mode-locked Cr2+:ZnSe laser. Reprinted with permission from Ma et al., Sci. Rep. 4, 5016 (2014). Copyright 2014 Nature Publishing Group.

Close modal

Table III summarizes the results of the passively mode-locked Cr2+:ZnSe and Cr2+:ZnS lasers. For each reference, we quote the shortest pulse width result.

TABLE III.

Passively mode-locked all-solid-state lasers operating in the 2.4 μm spectral region. ML is the mode-locking, τp is the pulse width, λ0 is the center wavelength, Pout is the average output power, NLM is the nonlinear mirror, EDFL is the Er3+-doped fiber laser, EYDFL is the Er3+/Yb3+-doped fiber laser, and TDFL is the Tm3+-doped fiber laser.

Laser gain materialML techniqueτp (fs)λ0 (nm)Pout (mW)Pump sourceReferences
Cr2+:ZnSe SESAM 10 800 ∼2500 400 Tm3+:YALO 218  
 SESAM 970 ∼2450 140 EDFL 219  
 SESAM 100 2450 75 EDFL 220  
 SESAM 80 2420 80 EDFL 221  
 SESAM 100 2450 100 EDFL 222  
 SESAM 132 2420 90 EDFL 224  
 KLM 95 2421 40 TDFL 225  
 KLM 100 2500 300 EDFL 226  
 KLM 100–130 2400 60 EDFL 233  
 KLM 92 2459 165 TDFL 227  
 Graphene 226 2500 80 TDFL 47  
 Graphene 116 2350 66 EYDFL 48  
 NLM 85 000 ∼2500 300 TDFL 236  
Cr2+:ZnS SESAM 1100 2450 125 EDFL 219  
 SESAM 110 2380 140 EDFL 223  
 SESAM 130 2375 130 EDFL 224  
 KLM 69 2390 550 EDFL 228  
 KLM 75 2390 1000 EDFL 230  
 KLM 68 2390 820 EDFL 229  
 KLM 125 2350 30 EDFL 233  
 KLM 41 ∼2400 1900 EDFL 234  
 KLM 29 ∼2400 440 EDFL 16  
 KLM 125 2347 80 Er3+:YAG 237  
 Graphene 189 2370 700 EDFL 49  
 Graphene 41 2400 ∼75 EDFL 50  
 Graphene 140 2350 1050 EDFL 51  
 Graphene 220 2327 880 EDFL 52  
 CNTs 61 2350 950 EDFL 41  
Laser gain materialML techniqueτp (fs)λ0 (nm)Pout (mW)Pump sourceReferences
Cr2+:ZnSe SESAM 10 800 ∼2500 400 Tm3+:YALO 218  
 SESAM 970 ∼2450 140 EDFL 219  
 SESAM 100 2450 75 EDFL 220  
 SESAM 80 2420 80 EDFL 221  
 SESAM 100 2450 100 EDFL 222  
 SESAM 132 2420 90 EDFL 224  
 KLM 95 2421 40 TDFL 225  
 KLM 100 2500 300 EDFL 226  
 KLM 100–130 2400 60 EDFL 233  
 KLM 92 2459 165 TDFL 227  
 Graphene 226 2500 80 TDFL 47  
 Graphene 116 2350 66 EYDFL 48  
 NLM 85 000 ∼2500 300 TDFL 236  
Cr2+:ZnS SESAM 1100 2450 125 EDFL 219  
 SESAM 110 2380 140 EDFL 223  
 SESAM 130 2375 130 EDFL 224  
 KLM 69 2390 550 EDFL 228  
 KLM 75 2390 1000 EDFL 230  
 KLM 68 2390 820 EDFL 229  
 KLM 125 2350 30 EDFL 233  
 KLM 41 ∼2400 1900 EDFL 234  
 KLM 29 ∼2400 440 EDFL 16  
 KLM 125 2347 80 Er3+:YAG 237  
 Graphene 189 2370 700 EDFL 49  
 Graphene 41 2400 ∼75 EDFL 50  
 Graphene 140 2350 1050 EDFL 51  
 Graphene 220 2327 880 EDFL 52  
 CNTs 61 2350 950 EDFL 41  

With the rapid development of mid-IR laser materials, particularly rare-earth-doped bulk gain media and fluoride fibers, laser emission in the 2.8 μm–3.5 μm spectral region has become a reality. Laser emission in this region is based primarily on optical transitions offered by Er3+-, Ho3+/Pr3+-, and Dy3+-doped gain materials, as shown in Fig. 8. Because the upper laser level has a shorter lifetime than the lower laser level in Er3+- and Ho3+-doped gain media, lasing processes are usually self-terminated. To overcome this limitation, several methods for achieving efficient laser operation have been explored, including energy transfer up-conversion, cascaded lasing, and co-doping with Pr3+ ions. Thus far, various all-solid-state and fiber laser systems have been demonstrated in the 2.8 μm–3.5 μm spectral region using Er3+, Ho3+, and Dy3+-doped gain materials.238–245 

FIG. 8.

Simplified energy level diagrams and energy transfer processes for (a) an Er3+-doped laser at 2.78 μm, (b) a Ho3+/Pr3+-co-doped laser at 2.85 μm, (c) a Dy3+-doped laser at 3.0 μm, and (d) an Er3+-doped laser at 3.5 μm. ETU: Energy transfer up-conversion.

FIG. 8.

Simplified energy level diagrams and energy transfer processes for (a) an Er3+-doped laser at 2.78 μm, (b) a Ho3+/Pr3+-co-doped laser at 2.85 μm, (c) a Dy3+-doped laser at 3.0 μm, and (d) an Er3+-doped laser at 3.5 μm. ETU: Energy transfer up-conversion.

Close modal

Owing to the water vapor absorption in the atmosphere, all-solid-state lasers that operate around 2.8 μm suffer considerable losses that make CW mode-locking formation particularly difficult. Thus far, no CW mode locking has been reported in all-solid-state lasers around 2.8 μm. Alternatively, fiber laser systems represent an ideal platform for ultrashort pulse generation in this region because the waveguide structure of fiber can eliminate the deleterious effect of intracavity water vapor absorption. Besides, fiber lasers are featured with high power, good beam quality, inherent simplicity and flexibility, and outstanding heat-dissipating capability. In 1996, Frerichs and Unrau reported the first Q-switched mode-locked fluoride fiber laser in the 3.0 μm spectral region by using flying-mirror technique and an InAs SA.246 However, low-quality fluoride fiber and an immature SA resulted in failure of realizing CW mode-locking operation. With the advances in fluoride fiber fabrication in recent years, commercial fluoride gain fibers are available with low loss (<50 dB/km), single-transverse-mode operation, and double-cladding structure, and different mid-IR SAs are developed, creating the conditions for ultrashort mid-IR pulse generation. Using a newly developed GaAs-SESAM, mode-locking operation was demonstrated in a Ho3+/Pr3+-doped ZBLAN fiber laser in 2012. The system delivered mode-locked pulses with a duration of 24 ps and a peak power of 0.2 kW at 2.87 μm.247 In the following years, mode-locked fluoride fiber lasers had been optimized continuously via adoption of various SAs, cavity structures, output couplers, and gain fiber types. However, the resulting pulses were still a few to tens of picoseconds in duration with low average power.248–250 In 2015, Tang et al. demonstrated a mode-locked Er3+-doped ZBLAN fiber laser at 2.8 μm (Fig. 9), with a record average output power of 1.05 W in single-pulse operation.251 This represented a significant milestone for CW mode-locked ZBLAN fiber lasers in the mid-IR region. The mode-locked pulses showed a pulse duration of 25 ps at 2780 nm. Since the Er3+-doped ZBLAN fiber possesses a broad emission bandwidth, wavelength-tunable mode-locking operation was achieved by incorporating a diffraction grating with a tunable range from 2710 nm to 2820 nm.252 

FIG. 9.

(a) Schematic of a high-average-power SESAM mode-locked Er3+-doped ZBLAN fiber laser. (b) Average output power vs incident pump power. (c) RF spectrum at the fundamental frequency. Inset: A wider-span RF spectrum. (d) Autocorrelation trace of the mode-locked pulses. (e) Output spectra of the CW and CW mode-locked Er3+-doped ZBLAN fiber laser. Reproduced with permission from Tang et al., Opt. Lett. 40, 4855 (2015). Copyright 2015 OSA Publishing.

FIG. 9.

(a) Schematic of a high-average-power SESAM mode-locked Er3+-doped ZBLAN fiber laser. (b) Average output power vs incident pump power. (c) RF spectrum at the fundamental frequency. Inset: A wider-span RF spectrum. (d) Autocorrelation trace of the mode-locked pulses. (e) Output spectra of the CW and CW mode-locked Er3+-doped ZBLAN fiber laser. Reproduced with permission from Tang et al., Opt. Lett. 40, 4855 (2015). Copyright 2015 OSA Publishing.

Close modal

Although various SAs such as Fe2+:ZnSe, InAs, and SESAM have been tested in Er3+- and Ho3+/Pr3+-doped fluoride fiber lasers, these mode-locked lasers typically operated in the picosecond regime. Femtosecond mode-locked fluoride fiber lasers were not achieved around 2.8 μm until NPR was employed. NPR mode-locked femtosecond fluoride fiber laser around 2.8 μm was first demonstrated In 2015.253,254Figure 10(a) shows the schematic of an NPR mode-locked Er:ZBLAN fluoride fiber laser, where 207 fs pulses were generated with a peak power of 3.4 kW at 2.8 μm [Fig. 10(b)]. Since fluoride fiber has an anomalous dispersion, the femtosecond mode-locked lasers generally operated in a soliton regime, as confirmed by the Kelly sidebands in the mode-locked spectrum [Fig. 10(c)]. Since the emission spectrum of the Er3+-doped fluoride fiber overlaps the absorption lines of water vapor, the mode-locked spectrum includes some absorption-related artifacts. This issue was mitigated by red shifting the laser wavelength toward 2.9 μm using a Ho3+/Pr3+-doped fluoride fiber. Antipov et al. reported an NPR mode-locked Ho3+/Pr3+-doped fluoride fiber laser in 2016. The laser produced 180 fs pulses with a peak power of 37 kW at 2876 nm.255 Recently, we also demonstrated an NPR mode-locked Er3+-doped fluoride fiber laser at 2.8 μm. The laser generated pulses with a duration of 131 fs by optimizing the gain fiber length, which are believed to be the shortest mode-locked pulses in the 2.8 μm wavelength region thus far.256 

FIG. 10.

(a) Schematic of the NPR mode-locked ring cavity. DM: dichroic mirror. M: gold-coated mirrors. ISO/POL: optical isolator also used as a polarizer. λ/4: quarter waveplate. λ/2: half-waveplate. CMS: cladding mode stripper. (b) Measured intensity autocorrelation curve fitted by a sech2(t)-shaped pulse with FWHM of 207 fs. (c) Measured output spectrum (blue), natural logarithm of the atmospheric transmission for a 1 m propagation inside the optical spectrum analyzer (black), reconstructed spectrum before the optical spectrum analyzer (red), and the theoretical spectrum of a sech2(t)-shaped pulse with FWHM of 230 fs (dashed gray). Reproduced with permission from Duval et al., Optica 2, 623 (2015). Copyright 2015 OSA Publishing.

FIG. 10.

(a) Schematic of the NPR mode-locked ring cavity. DM: dichroic mirror. M: gold-coated mirrors. ISO/POL: optical isolator also used as a polarizer. λ/4: quarter waveplate. λ/2: half-waveplate. CMS: cladding mode stripper. (b) Measured intensity autocorrelation curve fitted by a sech2(t)-shaped pulse with FWHM of 207 fs. (c) Measured output spectrum (blue), natural logarithm of the atmospheric transmission for a 1 m propagation inside the optical spectrum analyzer (black), reconstructed spectrum before the optical spectrum analyzer (red), and the theoretical spectrum of a sech2(t)-shaped pulse with FWHM of 230 fs (dashed gray). Reproduced with permission from Duval et al., Optica 2, 623 (2015). Copyright 2015 OSA Publishing.

Close modal

As previously mentioned, low-dimensional materials have been successfully used as SAs in mode-locked lasers that operate around 2.0 μm and 2.4 μm. Due to their broadband saturable absorption, ultrafast recovery time, and easy fabrication and integration, most of these materials have also been demonstrated as mode lockers for generation of ultrashort pulses at longer wavelength around 2.8 μm. Unlike most mode-locked all-solid-state lasers, fluoride fiber lasers operating around 2.8 μm generally require relatively large modulation depths for the SAs to initiate and sustain mode-locking operation. One effective way to increase the modulation depths of low-dimensional material SAs is to increase the sample thicknesses, such as by stacking single-layer graphene and directly synthesizing thicker TI samples. With multilayer graphene and Bi2Te3 nanosheets as SAs, CW mode-locking operation was achieved in fluoride fiber lasers around 2.8 μm, delivering mode-locked pulses with pulse durations of ∼42 ps and ∼6 ps, respectively.53,257 BP, which exhibits a layer-dependent bandgap that varies from ∼2.0 eV for monolayer to ∼0.3 eV for bulk, has also been demonstrated as a promising mid-IR saturable absorption material. In 2016, Qin et al. demonstrated the first BP mode-locked Er:ZBLAN fiber laser around 2.8 μm. It delivered 42 ps pulses with an average output power of 613 mW, as shown in Fig. 11.60 Subsequently, shorter 8.6 ps mode-locked pulses were achieved with BP as the mode locker in a Ho3+/Pr3+-doped fluoride fiber laser at 2866.7 nm.61 Cd3As2 film, a topological Dirac semimetal, has recently shown potential as a robust mid-IR SA for the development of ultrafast mid-IR laser sources. A mode-locked Ho3+/Pr3+-doped fluoride fiber laser was demonstrated with Cd3As2 film as SA, delivering 6.3 ps mode-locked pulses around 2.86 μm.63 Due to the immaturity of fluoride fiber-based optical devices and challenges in fluoride fiber splicing, most previous studies of mid-IR fluoride fiber lasers were performed using free-space optical elements that do not rival the advantages of all-fiber lasers with regard to flexible delivery, stability, and compactness. Recently, an all-fiber-designed fluoride fiber laser was constructed by taking advantage of BP flake integration and CW mode-locking operation was achieved at 2.8 μm, generating mode-locked pulses with estimated tens of picoseconds duration.258 

FIG. 11.

(a) Schematic of the BP mode-locked Er3+-doped 2.8 μm fiber laser. DM, dichroic mirror; ROC, radius of curvature; BP SAM, black phosphorus saturable absorber mirror. (b) Saturable absorption measurement of BP SAM at 2.8 μm. Inset: Experimental setup for saturable absorption measurement. (c) Autocorrelation trace of the mode-locked pulses (left) and the corresponding mode-locked pulse spectrum (right). Reproduced with permission from Qin et al., Opt. Lett. 41, 56 (2016). Copyright 2016 OSA Publishing.

FIG. 11.

(a) Schematic of the BP mode-locked Er3+-doped 2.8 μm fiber laser. DM, dichroic mirror; ROC, radius of curvature; BP SAM, black phosphorus saturable absorber mirror. (b) Saturable absorption measurement of BP SAM at 2.8 μm. Inset: Experimental setup for saturable absorption measurement. (c) Autocorrelation trace of the mode-locked pulses (left) and the corresponding mode-locked pulse spectrum (right). Reproduced with permission from Qin et al., Opt. Lett. 41, 56 (2016). Copyright 2016 OSA Publishing.

Close modal

The emission wavelengths of Er3+- and Ho3+/Pr3+-doped fluoride fibers are generally below 2.9 μm. This covers only a small part of the so-called molecular fingerprint region. There is a strong demand for compact, robust ultrafast laser sources at longer wavelengths where some important functional groups locate. One promising candidate is the Dy3+-doped fluoride fiber laser, which has a particularly broad tunable range from 2.8 μm to 3.4 μm.259 Mode locking of Dy3+-doped fluoride fiber laser was first demonstrated using frequency shifted feedback.260 This actively mode-locked laser delivered pulses with an average output power of 120 mW and a pulse duration of 33 ps at 3.1 μm. Based on the NPR mode locking technique, a femtosecond Dy3+-doped fluoride fiber laser was also demonstrated recently. The laser generated 724 fs transform-limited soliton pulses with a pulse energy of 2.3 nJ at 3.05 μm.261 Laser emission beyond 3.4 μm was also achieved using a dual-wavelength pumped Er3+-doped fluoride fiber laser, showing a wide tunable range from 3.4 μm to 3.8 μm.262 However, due to the lack of practical SAs in this region, the development of ultrafast laser was blocked for a long time. Until 2018, Qin et al. developed a bulk BP SAM and demonstrated CW mode locking in a dual-wavelength pumped Er3+-doped fluoride fiber laser at 3.5 μm (Fig. 12).17 The pulse duration was estimated to be tens of picoseconds according to the narrow spectral bandwidth of 4.7 nm. Femtosecond mode-locked fluoride fiber laser at 3.5 μm has not been achieved yet so far, and additional efforts need to be made due to shortage of mature optical devices around this wavelength region.

FIG. 12.

(a) Schematic of the BP mode-locked Er3+-doped fluoride fiber laser at 3.5 μm. OC: output coupler. CMS: cladding mode stripper. BP SAM: black phosphorus saturable absorber mirror. (b) Mode-locked pulse spectrum. (c) Mode-locked pulse train in microsecond and millisecond time scales. (d) Fundamental radio frequency spectrum with resolution bandwidth of 1 kHz. (e) Radio frequency spectrum with a broad span. Reproduced with permission from Qin et al., Opt. Express 26, 8224 (2018). Copyright 2018 OSA Publishing.

FIG. 12.

(a) Schematic of the BP mode-locked Er3+-doped fluoride fiber laser at 3.5 μm. OC: output coupler. CMS: cladding mode stripper. BP SAM: black phosphorus saturable absorber mirror. (b) Mode-locked pulse spectrum. (c) Mode-locked pulse train in microsecond and millisecond time scales. (d) Fundamental radio frequency spectrum with resolution bandwidth of 1 kHz. (e) Radio frequency spectrum with a broad span. Reproduced with permission from Qin et al., Opt. Express 26, 8224 (2018). Copyright 2018 OSA Publishing.

Close modal

Table IV summarizes the results of CW mode-locked Er3+-, Ho3+/Pr3+-, and Dy3+-doped fluoride fiber lasers in the 2.8 μm–3.5 μm spectral region. For each reference, we quote the shortest pulse duration result.

TABLE IV.

Mode-locked Er3+-, Ho3+/Pr3+-, and Dy3+-doped fluoride fiber lasers in the 2.8 μm–3.5 μm spectral region. ML is the mode-locking, τp is the pulse width, λ0 is the center wavelength, Pout is the average output power, FSF is the frequency-shifted feedback, NLM is the nonlinear mirror, and AOM is the acousto-optic modulator.

Laser gain materialML techniqueτp (ps)λ0 (nm)Pout (mW)Ppeak (W)References
Er3+ fluoride fiber Fe:ZnSe ∼19 2783 51.4 49 248  
 SESAM 25 2780 1050 1900 251  
 Graphene 42 2784.5 18 17 53  
 SESAM 60 2797 440 140 249  
 NPR 0.207 2805 44 3500 253  
 NPR 0.497 2793 206 6400 254  
 NPR 0.131 2780 317 23 000 256  
 BP 42 2783 613 610 60  
 BP 2771 6.2 … 258  
 NLM ∼2800 88 … 263  
Ho3+/Pr3+ fluoride fiber SESAM 24 2870 132 200 247  
 AOM 2870 34 330 264  
 InAs SA ∼2860 69.2 470 250  
 SESAM 6.4 2710–2820 203 1100 252  
 Bi2Te3 ∼6 2830 90 1400 257  
 BP 8.6 2866.7 87.8 730 61  
 NPR 0.180 2876 327 37 000 255  
 Cd3As2 film 6.3 2860 … … 63  
Dy3+ fluoride fiber FSF 33 2970–3300 120 82 260  
 NPR 0.724 3050 175 3110 261  
Er3+ fluoride fiber BP … 3462 40 … 17  
Laser gain materialML techniqueτp (ps)λ0 (nm)Pout (mW)Ppeak (W)References
Er3+ fluoride fiber Fe:ZnSe ∼19 2783 51.4 49 248  
 SESAM 25 2780 1050 1900 251  
 Graphene 42 2784.5 18 17 53  
 SESAM 60 2797 440 140 249  
 NPR 0.207 2805 44 3500 253  
 NPR 0.497 2793 206 6400 254  
 NPR 0.131 2780 317 23 000 256  
 BP 42 2783 613 610 60  
 BP 2771 6.2 … 258  
 NLM ∼2800 88 … 263  
Ho3+/Pr3+ fluoride fiber SESAM 24 2870 132 200 247  
 AOM 2870 34 330 264  
 InAs SA ∼2860 69.2 470 250  
 SESAM 6.4 2710–2820 203 1100 252  
 Bi2Te3 ∼6 2830 90 1400 257  
 BP 8.6 2866.7 87.8 730 61  
 NPR 0.180 2876 327 37 000 255  
 Cd3As2 film 6.3 2860 … … 63  
Dy3+ fluoride fiber FSF 33 2970–3300 120 82 260  
 NPR 0.724 3050 175 3110 261  
Er3+ fluoride fiber BP … 3462 40 … 17  

Since many important molecules have their characteristic vibration and rotation spectra in the midinfrared region, ultrafast midinfrared lasers provide new opportunities for a host of application areas such as molecular spectroscopy, chemical and biomolecular sensing, etc. In addition, ultrafast midinfrared sources have also been proved as a powerful tool to generate midinfrared supercontinuum and laser frequency combs, and to investigate ultrafast dynamic processes in solid-state physics. In this section, we will highlight some representative applications of ultrashort mid-IR mode-locked laser sources.

Supercontinuum source, featured with an extremely broad emission bandwidth, is considered a very important tool for many applications and attracts a growing attention in the mid-IR region. Although supercontinuum generation technology is well developed and has even reached commercialization in the visible and near-IR spectral regions, it remains challenging to provide high-efficiency, high-power mid-IR supercontinuum sources. The limitations concern both highly transparent nonlinear media in the longer mid-IR region and reliable ultrafast pump sources that can be used to pump nonlinear media efficiently around zero-dispersion wavelengths. Benefiting from the development of ultrafast mid-IR mode-locked lasers, remarkable progress has been made in supercontinuum generation in recent years. Pumping by a CNT mode-locked Tm3+-doped fiber laser around 2.0 μm, Thapa et al. reported ultrawide mid-IR supercontinuum generation spanning from 0.9 μm to 4.2 μm in a low-loss W-type tellurite fiber.265 They successfully developed fiber fusion splicing technology capable of splicing tellurite fibers directly to silica fibers, paving the way for all-fiber mid-IR supercontinuum sources. Compared to tellurite glass with a transmission window limited to ∼5 μm, chalcogenide materials such as As2Se3 and GaSe not only exhibit much broader transmission windows up to 15 μm and even into the low-frequency terahertz range266–268 but also have much higher nonlinear refractive indices. This makes them the most promising candidates for broadband mid-IR supercontinuum generation. By combining a newly developed mode-locked Ho3+/Pr3+-doped ZBLAN fiber laser with a dispersion-engineered As2Se3/As2S3 tapered fiber device, Hudson et al. achieved a supercontinuum with a bandwidth of over 10 000 nm (125 THz) in the mid-IR region,269 as shown in Fig. 13. The pump laser emitted 230 fs pulses around 2.9 μm at a repetition rate of 42 MHz with an average power of 140 mW. It generated a 2.4 octave-spanning supercontinuum covering from ∼2 μm to 12 μm range. More recently, Zhang et al. demonstrated broadband, coherent, and carrier-envelope-phase-stable supercontinuum radiation from a 1 mm-thick GaSe crystal driven by ultrashort self-compressed pulses from a KLM Ho3+:YAG thin-disk oscillator.270 The generated supercontinuum had a spectrum of 500 cm−1–2250 cm−1 (−30 dB), which corresponded to a wavelength range from 4.5 μm to 20 μm. This suggests excellent potential for future spectroscopy applications.

FIG. 13.

(a) Layout of the mode-locked Ho3+/Pr3+-doped ZBLAN fiber laser cavity and the subsequent supercontinuum generation stage. Inset: Dispersion of both the untapered step-index fiber and the microwire section. (b) Spectral expansion of supercontinuum at different pump peak powers. Reproduced with permission from Hudson et al., Optica 4, 1163 (2017). Copyright 2017 OSA Publishing.

FIG. 13.

(a) Layout of the mode-locked Ho3+/Pr3+-doped ZBLAN fiber laser cavity and the subsequent supercontinuum generation stage. Inset: Dispersion of both the untapered step-index fiber and the microwire section. (b) Spectral expansion of supercontinuum at different pump peak powers. Reproduced with permission from Hudson et al., Optica 4, 1163 (2017). Copyright 2017 OSA Publishing.

Close modal

Laser frequency combs have dramatically improved precise measurements of frequency and time. They have been used widely in metrology and physics research since their introduction in the late 1990s. In the mid-IR range, optical frequency combs are of significant interest for molecular spectroscopy. They are expected to enable ultrasensitive, rapid, precise molecular sensing across the broadband mid-IR molecular fingerprint region. Although several approaches to mid-IR frequency comb generation have previously been reported,271–274 mid-IR mode-locked laser sources are probably the most general, reliable, naturally low-noise, and compact choices for frequency comb generation. Using SESAM as a mode locker, Bernhardt et al. presented frequency-comb Fourier-transform spectroscopy based on two ceramic Cr2+:ZnSe femtosecond oscillators emitting in the 2.4 μm mid-IR region.275 Ultrashort pulses as short as ∼3 optical cycles have been recently reported from a Cr2+:ZnS mode-locked laser, emitting a very broadband spectrum around 2.5 μm.16 This unique mid-IR mode-locked laser generated intrinsically coherent frequency combs, further extending potential applications to high-resolution, high-sensitivity spectroscopy and optical clocks. In the 2.0 μm–3.5 μm wavelength region, mode-locked lasers have also been achieved using a variety of techniques based on Tm3+, Ho3+, Er3+, and Dy3+-doped materials, which may directly extend the spectrum coverage region and offer potential sources for laser frequency combs in the mid-IR region.271,276

The development of ultrafast mid-IR laser sources significantly advances progress in mid-IR spectroscopy, especially ultrafast mid-IR spectroscopy, which are very important sample characterization and analytical techniques in the fields of molecular science and solid-state physics.277–282 They have been used extensively in applications that involve qualitative and quantitative analyses, investigating the structural information of molecules and its governing physical laws, and providing the time-dependent physical parameters for dynamic processes investigations. For example, structural dynamics of molecules such as vibrational mode coupling and charge transfer were investigated via time-resolved mid-IR spectroscopy at the femtosecond time scale.279,283 Free-carrier dynamics were clearly observed in semiconductors via ultrafast mid-IR pump-probe spectroscopy.284,285 Furthermore, two-dimensional infrared spectroscopy, a novel analytical technique based on time-resolved detection of mid-IR signals, has also been developed benefitting from the ability to generate ultrashort mid-IR pulses.277,280,286,287 Two-dimensional infrared spectroscopy can provide direct insight into structural dynamics such as vibrational mode coupling, anharmonicities, bond formation and dissociation, and hydrogen network alternation at the femtosecond time scale, offering much more precise information than linear mid-IR spectroscopy. As mentioned in Sec. III, compared to ultrafast sources generated by nonlinear processes, mid-IR mode-locked lasers have the advantages of compactness, reliability, low cost, and the ability to directly generate ultrashort femtosecond pulses with good stability and beam quality. Using a femtosecond Cr2+:ZnSe laser, Sorokin et al. demonstrated mid-IR spectroscopy for gas detection and concentration analysis, as shown in Fig. 14.282 They achieved high sensitivity and resolution over a broad spectral domain. In addition, with mid-IR mode-locked laser sources, one can extend the spectrum to longer mid-IR regions and generate pulses as short as a few cycles via nonlinear techniques.269 This has the potential to further aid ultrafast mid-IR spectroscopy.

FIG. 14.

Experimental setup of mode-locked Cr2+:ZnSe all-solid-state laser. OC: output coupler. Atten: attenuator. (b) Part of a C2H2 spectrum at 0.12 cm−1 resolution in the 2.4 μm region. Inset: Acetylene and residual water vapor lines coming from a 5 m air path between the cell and the interferometer. Reproduced with permission from Sorokin et al., Opt. Express 15, 16540 (2007). Copyright 2007 OSA Publishing.

FIG. 14.

Experimental setup of mode-locked Cr2+:ZnSe all-solid-state laser. OC: output coupler. Atten: attenuator. (b) Part of a C2H2 spectrum at 0.12 cm−1 resolution in the 2.4 μm region. Inset: Acetylene and residual water vapor lines coming from a 5 m air path between the cell and the interferometer. Reproduced with permission from Sorokin et al., Opt. Express 15, 16540 (2007). Copyright 2007 OSA Publishing.

Close modal

Material processing using ultrashort mid-IR laser pulses has also attracted significant interest in recent years due to its potential applications in science, technology, and industry.288–292 Ultrafast mid-IR laser sources extend existing ultrafast laser processing technology into the mid-IR region. This opens the possibility of expanding high-precision processing techniques to materials (e.g., silicon, germanium) that are opaque in the visible and near-infrared spectral ranges but optically transparent in the mid-IR region. Employing a femtosecond laser source at 2.4 μm, Nejadmalayeri et al. fabricated a single-mode optical waveguide for the 1.55 μm and 1.32 μm telecommunications bands in bulk crystalline silicon via ultrafast mid-IR laser inscription.288 The waveguide exhibited a low insertion loss of ∼1.2 dB/cm or less, which made it applicable to silicon-based optical circuits.293 With further optimization, laser inscription with ultrashort mid-IR laser pulses could be applied to in-chip microstructure fabrication, where it could be used to create complex three-dimensional optical circuits directly inside silicon, overcoming problems that current lithography methods cannot solve. Recently, a numerical study of three-photon absorption vs two-photon absorption for in-bulk modification of silicon indicated that mid-IR femtosecond lasers have the potential to be used for efficient kerfless silicon processing.291 This technique can be employed to replace the currently used diamond saws that separate wafers from the silicon bulk and minimize wasted material. Polymer processing, including cutting and welding, was also investigated using ultrashort mid-IR laser pulses. Using an all-fiber ultrafast mid-IR laser around 2 μm, two 5 mm-thick polymethyl methacrylate blocks were butt-welded with a stable joint, and Kapton film was cut with a cut width of only 11 μm.292 

In the medical field, ultrafast mid-IR laser sources have been demonstrated to be powerful tools and shown great advantages for the applications of laser surgery and biodiagnostics.294–304 The mid-IR wavelength around 3.0 μm corresponds to the vibrational absorption band of water molecules within tissue. Thus, laser radiation can be substantially absorbed within a few microns of incidence. Furthermore, the picosecond-level pulse duration is short enough to drive tissue ablation faster than the thermal and acoustic energy transfer processes that cause damage to the surrounding tissue, but long enough to avoid tissue ionization and the risks of creating toxic free radicals and fragmenting constituent proteins.294,295,305 Therefore, high-energy mid-IR picosecond laser pulses around 3 μm are of particular interest for minimally invasive surgery and may be better than conventional medical lasers with pulse durations longer than nanoseconds. Figure 15 shows schematics of three cutting modalities by using a mid-IR picosecond laser system, a conventional long-pulsed surgical laser, and a mechanical surgical scalpel.295 The mechanical surgical scalpel causes more than 400 μm damage zone from the borders of the incision and the conventional laser damages skin border up to 800 μm away from the ablated edge. In contrast, cuts using the mid-IR picosecond laser system have sharp edges and a damage zone of less than 10 μm, which is even smaller than the diameter of a single skin fibroblast. In addition, an investigation of wound healing has demonstrated that wounds produced by a mid-IR picosecond laser healed faster than those produced via either a conventional surgical laser or a mechanical surgical scalpel. The scars formed were half the size of those from the aforementioned conventional methods. These advantages can decrease the risk of infection after surgery and increase patient comfort. Aside from promising applications as surgery tools, ultrafast mid-IR laser sources have significant potential for early real-time biodiagnostics.302,304 Mouras et al. recently used a high-repetition-rate (80 MHz), tunable ultrafast mid-IR laser to develop a high-speed, chemical selective, label-free bench-top mid-IR microscope for chemical imaging.304 This microscope had a subwavelength resolution and could be applied to living cells and tissue, where it offered real-time tracking of drugs and drug carriers within cells and tissues.

FIG. 15.

Schematics of cutting modalities. (a) The mechanical scalpel cuts skin by producing shear forces which exceed the elastic limit of the tissue. This causes a border of damage around the incision which reaches as far as 400 μm from the borders of the incision. (b) Conventional surgical lasers cut by depositing heat until the tissue melts or burns away. The damage zone in this case can reach up to 800 μm away from the ablated edge. (c) By contrast, the well absorbed picosecond IR laser pulses cause superheating of water inside the tissue on the picosecond time scale, ejecting the tissue faster than energy can diffuse to the surroundings area. The remaining adjacent tissue shows minimal damage compared to the other two modalities. Reproduced with permission from Amini-Nik et al., PLoS One 5, e13053 (2010). Copyright 2010 Public Library of Science.

FIG. 15.

Schematics of cutting modalities. (a) The mechanical scalpel cuts skin by producing shear forces which exceed the elastic limit of the tissue. This causes a border of damage around the incision which reaches as far as 400 μm from the borders of the incision. (b) Conventional surgical lasers cut by depositing heat until the tissue melts or burns away. The damage zone in this case can reach up to 800 μm away from the ablated edge. (c) By contrast, the well absorbed picosecond IR laser pulses cause superheating of water inside the tissue on the picosecond time scale, ejecting the tissue faster than energy can diffuse to the surroundings area. The remaining adjacent tissue shows minimal damage compared to the other two modalities. Reproduced with permission from Amini-Nik et al., PLoS One 5, e13053 (2010). Copyright 2010 Public Library of Science.

Close modal

The development of robust, compact, cost-effective ultrafast mid-IR laser sources with advanced performances has a rich history. Passively mode-locked lasers in the mid-IR region act as ideal sources as they can generally satisfy these requirements and provide high power, ultrashort pulses, and good beam quality. However, there was little progress in the development of passively mode-locked mid-IR laser sources until the recent decade. Driven by the rapid development of mid-IR laser gain materials and mode locking techniques, significant advances were made in ultrafast mode-locked lasers for the 2.0 μm–3.5 μm mid-IR region by using various rare-earth-ion (Tm3+, Ho3+, Er3+, and Dy3+)-doped solid-state and fiber materials and transition-metal Cr2+-doped ZnS and ZnSe gain media. Remarkably, pulse duration from mid-IR mode-locked laser has reached down to ∼3 optical cycles in the Cr2+:ZnS laser, and the average output power from a mid-IR mode-locked laser has been as high as 20 W from a Ho3+:YAG laser. The wavelength range, which is particularly important for molecular spectroscopy, has also been extended to the 3.5 μm region. The mid-IR mode-locked laser achievements summarized and highlighted in this review represent crucial and recent advances in this research field. Furthermore, these significant progresses on mid-IR ultrashort laser sources open the door for a series of applications including mid-IR supercontinuum generation, mid-IR frequency combs, mid-IR spectroscopy, material processing, laser surgery, and biodiagnostics.

In the meantime, although significant progress has been achieved in recent years, developing compact and robust ultrafast mid-IR laser sources with better performances such as broader wavelength range covering 5–10 μm, shorter pulse duration down to one optical cycle, higher pulse energy, and peak power levels that can reach the strong-field regime, is still on the way. Several challenges remain to be addressed. First, the development of mid-IR mode-locked laser sources has historically been hindered by poor laser material availability. Thus, generation of ultrashort mid-IR pulses via nonlinear frequency conversion has become a general method, especially in the long wavelength region that current gain materials cannot support. To realize direct ultrashort laser generation with high efficiency and high power, novel gain media that can operate in the longer mid-IR wavelengths need to be developed. Meanwhile, few-cycle ultrashort mid-IR pulse generation and amplification have become increasingly attractive in recent years, which depend on the development of novel broadband gain materials with smooth, flat gain profiles and excellent mechanical, thermal, and optical properties. Second, significant advances have been made in the mid-IR mode-locking devices and techniques in recent years which are the critical factors influencing the mode-locked laser performance. However, compared to the mature SAs available for the near-IR region, there is still significant room for further improvement and optimization of current mid-IR SAs in terms of recovery time, nonsaturable loss, and damage threshold. Considerable efforts are still required for development of novel saturable absorption materials and mode-locked techniques for ultrashort mid-IR pulse generation. In addition, mid-IR SAs based on low-dimensional materials are in their infancy. Exploring reliable approaches to reproducible SA fabrication with precisely customized parameters is still an important task in the future work. Third, ultrashort mid-IR laser sources have become of interest to strong-field physics, as strong laser-matter interactions are dependent on the laser wavelength (high-harmonic generation cutoff frequency ∝ λ2, attosecond pulse duration ∝ λ−1/2, electron kinetic energies ∝ λ2).306–308 Thus, developing novel ultrashort mid-IR pulse generation and amplification technologies that can achieve high energy and peak power levels is increasingly important for strong field science. Although some preliminary work such as development of Cr2+:ZnSe regenerative amplifiers309 and Cr2+:ZnSe thin disk lasers310 has been performed, a lot of work remains to be done for achieving robust, compact, and practical tabletop ultrashort mid-IR laser sources with high energy and peak power levels, such as development of mid-IR dispersion devices, exploration of mid-IR pulse and spectrum shaping techniques, realization of carrier envelope phase control, etc. With continued development of novel materials and technologies in the mid-IR region, we believe that ultrashort mid-IR pulses and even few-cycle pulses with high average power, high pulse energy, and longer operation wavelength can be expected in the future. Breakthroughs in ultrafast mid-IR laser sources will open new research avenues and lead to broad applications in science, industry, medicine, and other fields.

This research was partially supported by funds from the National Natural Science Foundation of China (Grant Nos. 61575089, 61675130, and 11721091), the Jiangsu Specially-Appointed Professor Program, the National Postdoctoral Program for Innovative Talents (Grant No. BX20170149), the China Postdoctoral Science Foundation (Grant No. 2017M620150), and A*Star Singapore (Grant No. A1883c0003).

1.
R. R.
Gattass
and
E.
Mazur
,
Nat. Photonics
2
,
219
(
2008
).
2.
M.
Ebrahim-Zadeh
and
K.
Vodopyanov
,
J. Opt. Soc. Am. B
33
,
MIC1
(
2016
).
3.
J.
Mandon
,
G.
Guelachvili
, and
N.
Picque
,
Nat. Photonics
3
,
99
(
2009
).
4.
P.
Loza-Alvarez
,
C. T. A.
Brown
,
D. T.
Reid
,
W.
Sibbett
, and
M.
Missey
,
Opt. Lett.
24
,
1523
(
1999
).
5.
K.
Zhao
,
H.
Zhong
,
P.
Yuan
,
G.
Xie
,
J.
Wang
,
J.
Ma
, and
L.
Qian
,
Opt. Lett.
38
,
2159
(
2013
).
6.
G.
Andriukaitis
,
T.
Balčiūnas
,
S.
Ališauskas
,
A.
Pugžlys
,
A.
Baltuška
,
T.
Popmintchev
,
M.-C.
Chen
,
M. M.
Murnane
, and
H. C.
Kapteyn
,
Opt. Lett.
36
,
2755
(
2011
).
7.
H.
Okamoto
and
M.
Tasumi
,
Opt. Commun.
121
,
63
(
1995
).
8.
T.
Flöry
,
P.
Malevich
,
A.
Pugžlys
,
A.
Voronin
,
A. M.
Zheltikov
, and
A.
Baltuška
, “
Generation of ultra-short mid-IR by four-wave mixing in metalized waveguides
,” in
CLEO/Europe 2015, OSA Technical Digest
(Optical Society of America,
2015
), Paper No. CD_4_5.
9.
T.
Fuji
and
T.
Suzuki
,
Opt. Lett.
32
,
3330
(
2007
).
11.
S.
Mirov
,
V.
Fedorov
,
I.
Moskalev
,
D.
Martyshkin
, and
C.
Kim
,
Laser Photonics Rev.
4
,
21
(
2010
).
12.
X.
Zhu
and
N.
Peyghambarian
,
Adv. OptoElectron.
2010
,
501956
.
13.
S. D.
Jackson
,
Nat. Photonics
6
,
423
(
2012
).
14.
J.
Hu
,
J.
Meyer
,
K.
Richardson
, and
L.
Shah
,
Opt. Mater. Express
3
,
1571
(
2013
).
15.
Y.
Wang
,
W.
Jing
,
P.
Loiko
,
Y.
Zhao
,
H.
Huang
,
S.
Suomalainen
,
A.
Härkönen
,
M.
Guina
,
X.
Mateos
,
U.
Griebner
, and
V.
Petrov
, “
Sub-10 optical-cycle mode-locked Tm:(Lu2/3Sc1/3)2O3 mixed ceramic laser at 2057 nm
,” in
ASSL 2017, OSA Technical Digest
(Optical Society of America,
2017
), Paper No. ATu6A.4.
16.
S.
Vasilyev
,
I.
Moskalev
,
M.
Mirov
,
S.
Mirov
, and
V.
Gapontsev
,
Opt. Lett.
40
,
5054
(
2015
).
17.
Z.
Qin
,
T.
Hai
,
G.
Xie
,
J.
Ma
,
P.
Yuan
,
L.
Qian
,
L.
Li
,
L.
Zhao
, and
D.
Shen
,
Opt. Express
26
,
8224
(
2018
).
18.
U.
Keller
,
D. A. B.
Miller
,
G. D.
Boyd
,
T. H.
Chiu
,
J. F.
Ferguson
, and
M. T.
Asom
,
Opt. Lett.
17
,
505
(
1992
).
19.
K.
Yang
,
D.
Heinecke
,
J.
Paajaste
,
C.
Kölbl
,
T.
Dekorsy
,
S.
Suomalainen
, and
M.
Guina
,
Opt. Express
21
,
4311
(
2013
).
20.
A. A.
Lagatsky
,
F.
Fusari
,
S.
Calvez
,
S. V.
Kurilchik
,
V. E.
Kisel
,
N. V.
Kuleshov
,
M. D.
Dawson
,
C. T. A.
Brown
, and
W.
Sibbett
,
Opt. Lett.
35
,
172
(
2010
).
21.
V.
Aleksandrov
,
A.
Gluth
,
V.
Petrov
,
I.
Buchvarov
,
G.
Steinmeyer
,
J.
Paajaste
,
S.
Suomalainen
,
A.
Härkönen
,
M.
Guina
,
X.
Mateos
,
F.
Díaz
, and
U.
Griebner
,
Opt. Express
23
,
4614
(
2015
).
22.
J.
Paajaste
,
S.
Suomalainen
,
R.
Koskinen
,
A.
Härkönen
,
G.
Steinmeyer
, and
M.
Guina
,
Phys. Status Solidi C
9
,
294
(
2012
).
23.
A. A.
Lagatsky
,
F.
Fusari
,
S.
Calvez
,
J. A.
Gupta
,
V. E.
Kisel
,
N. V.
Kuleshov
,
C. T. A.
Brown
,
M. D.
Dawson
, and
W.
Sibbett
,
Opt. Lett.
34
,
2587
(
2009
).
24.
A. A.
Lagatsky
,
X.
Han
,
M. D.
Serrano
,
C.
Cascales
,
C.
Zaldo
,
S.
Calvez
,
M. D.
Dawson
,
J. A.
Gupta
,
C. T. A.
Brown
, and
W.
Sibbett
,
Opt. Lett.
35
,
3027
(
2010
).
25.
A. A.
Lagatsky
,
S.
Calvez
,
J. A.
Gupta
,
V. E.
Kisel
,
N. V.
Kuleshov
,
C. T. A.
Brown
,
M. D.
Dawson
, and
W.
Sibbett
,
Opt. Express
19
,
9995
(
2011
).
26.
A. A.
Lagatsky
,
O. L.
Antipov
, and
W.
Sibbett
,
Opt. Express
20
,
19349
(
2012
).
27.
A.
Gluth
,
Y.
Wang
,
V.
Petrov
,
J.
Paajaste
,
S.
Suomalainen
,
A.
Härkönen
,
M.
Guina
,
G.
Steinmeyer
,
X.
Mateos
,
S.
Veronesi
,
M.
Tonelli
,
J.
Li
,
Y.
Pan
,
J.
Guo
, and
U.
Griebner
,
Opt. Express
23
,
1361
(
2015
).
28.
Y.
Wang
,
G.
Xie
,
X.
Xu
,
J.
Di
,
Z.
Qin
,
S.
Suomalainen
,
M.
Guina
,
A.
Härkönen
,
A.
Agnesi
,
U.
Griebner
,
X.
Mateos
,
P.
Loiko
, and
V.
Petrov
,
Opt. Mater. Express
6
,
131
(
2016
).
29.
A. A.
Lagatsky
,
P.
Koopmann
,
P.
Fuhrberg
,
G.
Huber
,
C. T. A.
Brown
, and
W.
Sibbett
,
Opt. Lett.
37
,
437
(
2012
).
30.
A.
Härkönen
,
J.
Paajaste
,
S.
Suomalainen
,
J.-P.
Alanko
,
C.
Grebing
,
R.
Koskinen
,
G.
Steinmeyer
, and
M.
Guina
,
Opt. Lett.
35
,
4090
(
2010
).
31.
C.
Luan
,
K.
Yang
,
J.
Zhao
,
S.
Zhao
,
T.
Li
,
H.
Zhang
,
J.
He
,
L.
Song
,
T.
Dekorsy
,
M.
Guina
, and
L.
Zheng
,
Opt. Lett.
42
,
839
(
2017
).
32.
C.
Lin
,
M.
Grau
,
O.
Dier
, and
M. C.
Amann
,
Appl. Phys. Lett.
84
,
5088
(
2004
).
33.
W. B.
Cho
,
A.
Schmidt
,
J. H.
Yim
,
S. Y.
Choi
,
S.
Lee
,
F.
Rotermund
,
U.
Griebner
,
G.
Steinmeyer
,
V.
Petrov
,
X.
Mateos
,
M. C.
Pujol
,
J. J.
Carvajal
,
M.
Aguiló
, and
F.
Díaz
,
Opt. Express
17
,
11007
(
2009
).
34.
A.
Schmidt
,
P.
Koopmann
,
G.
Huber
,
P.
Fuhrberg
,
S. Y.
Choi
,
D.-I.
Yeom
,
F.
Rotermund
,
V.
Petrov
, and
U.
Griebner
,
Opt. Express
20
,
5313
(
2012
).
35.
A.
Schmidt
,
S. Y.
Choi
,
D.-I.
Yeom
,
F.
Rotermund
,
X.
Mateos
,
M.
Segura
,
F.
Diaz
,
V.
Petrov
, and
U.
Griebner
,
Appl. Phys. Express
5
,
092704
(
2012
).
36.
V.
Aleksandrov
,
A.
Gluth
,
V.
Petrov
,
I.
Buchvarov
,
S. Y.
Choi
,
M. H.
Kim
,
F.
Rotermund
,
X.
Mateos
,
F.
Díaz
, and
U.
Griebner
,
Opt. Express
22
,
26872
(
2014
).
37.
Z.
Pan
,
Y.
Wang
,
Y.
Zhao
,
M.
Kowalczyk
,
J.
Sotor
,
H.
Yuan
,
Y.
Zhang
,
X.
Dai
,
H.
Cai
,
J. E.
Bae
,
S. Y.
Choi
,
F.
Rotermund
,
P.
Loiko
,
J. M.
Serres
,
X.
Mateos
,
U.
Griebner
, and
V.
Petrov
,
Opt. Lett.
43
,
5154
(
2018
).
38.
Z.
Pan
,
Y.
Wang
,
Y.
Zhao
,
H.
Yuan
,
X.
Dai
,
H.
Cai
,
J. E.
Bae
,
S. Y.
Choi
,
F.
Rotermund
,
X.
Mateos
,
J. M.
Serres
,
P.
Loiko
,
U.
Griebner
, and
V.
Petrov
,
Photonics Res.
6
,
800
(
2018
).
39.
Y.
Wang
,
Y.
Zhao
,
Z.
Pan
,
J. E.
Bae
,
S. Y.
Choi
,
F.
Rotermund
,
P.
Loiko
,
J. M.
Serres
,
X.
Mateos
,
H.
Yu
,
H.
Zhang
,
M.
Mero
,
U.
Griebner
, and
V.
Petrov
,
Opt. Lett.
43
,
4268
(
2018
).
40.
Y.
Zhao
,
Y.
Wang
,
X.
Zhang
,
X.
Mateos
,
Z.
Pan
,
P.
Loiko
,
W.
Zhou
,
X.
Xu
,
J.
Xu
,
D.
Shen
,
S.
Suomalainen
,
A.
Härkönen
,
M.
Guina
,
U.
Griebner
, and
V.
Petrov
,
Opt. Lett.
43
,
915
(
2018
).
41.
N.
Tolstik
,
O.
Okhotnikov
,
E.
Sorokin
, and
I. T.
Sorokina
,
Proc. SPIE
8959
,
89591A
(
2014
).
42.
J.
Ma
,
G. Q.
Xie
,
P.
Lv
,
W. L.
Gao
,
P.
Yuan
,
L. J.
Qian
,
H. H.
Yu
,
H. J.
Zhang
,
J. Y.
Wang
, and
D. Y.
Tang
,
Opt. Lett.
37
,
2085
(
2012
).
43.
G. Q.
Xie
,
J.
Ma
,
P.
Lv
,
W. L.
Gao
,
P.
Yuan
,
L. J.
Qian
,
H. H.
Yu
,
H. J.
Zhang
,
J. Y.
Wang
, and
D. Y.
Tang
,
Opt. Mater. Express
2
,
878
(
2012
).
44.
A. A.
Lagatsky
,
Z.
Sun
,
T. S.
Kulmala
,
R. S.
Sundaram
,
S.
Milana
,
F.
Torrisi
,
O. L.
Antipov
,
Y.
Lee
,
J. H.
Ahn
,
C. T. A.
Brown
,
W.
Sibbett
, and
A. C.
Ferrari
,
Appl. Phys. Lett.
102
,
013113
(
2013
).
45.
B.
Yang
,
Q.
Mei
,
W.
Shuang
,
S.
Ruiping
,
L.
Diao
,
R.
Zhaoyu
, and
B.
Jintao
,
Appl. Phys. Express
6
,
102701
(
2013
).
46.
M.
Jie
,
X.
Guoqiang
,
Z.
Jian
,
Y.
Peng
,
T.
Dingyuan
, and
Q.
Liejia
,
IEEE J. Sel. Top. Quantum Electron.
21
,
50
(
2015
).
47.
M. N.
Cizmeciyan
,
J. W.
Kim
,
S.
Bae
,
B. H.
Hong
,
F.
Rotermund
, and
A.
Sennaroglu
,
Opt. Lett.
38
,
341
(
2013
).
48.
J.
Ma
,
G.
Xie
,
P.
Lv
,
W.
Gao
,
P.
Yuan
,
L.
Qian
,
U.
Griebner
,
V.
Petrov
,
H.
Yu
,
H.
Zhang
, and
J.
Wang
,
Sci. Rep.
4
,
5016
(
2014
).
49.
N.
Tolstik
,
A.
Pospischil
,
E.
Sorokin
, and
I. T.
Sorokina
,
Opt. Express
22
,
7284
(
2014
).
50.
N.
Tolstik
,
E.
Sorokin
, and
I. T.
Sorokina
,
Opt. Express
22
,
5564
(
2014
).
51.
N.
Tolstik
,
E.
Sorokin
, and
I. T.
Sorokina
, “
Ceramic Cr:ZnS laser mode-locked by graphene
,” in
CLEO/S&I 2014, OSA Technical Digest
(Optical Society of America,
2014
), Paper No. STu2E.7.
52.
W. B.
Cho
,
S. Y.
Choi
,
C.
Zhu
,
M. H.
Kim
,
J. W.
Kim
,
J. S.
Kim
,
H. J.
Park
,
D. H.
Shin
,
M. Y.
Jung
,
F.
Wang
, and
F.
Rotermund
,
Opt. Express
24
,
20774
(
2016
).
53.
G.
Zhu
,
X.
Zhu
,
F.
Wang
,
S.
Xu
,
Y.
Li
,
X.
Guo
,
K.
Balakrishnan
,
R. A.
Norwood
, and
N.
Peyghambarian
,
IEEE Photonics Technol. Lett.
28
,
7
(
2016
).
54.
C.
Zhao
,
H.
Zhang
,
X.
Qi
,
Y.
Chen
,
Z.
Wang
,
S.
Wen
, and
D.
Tang
,
Appl. Phys. Lett.
101
,
211106
(
2012
).
55.
J.
Sotor
,
G.
Sobon
, and
K. M.
Abramski
,
Opt. Express
22
,
13244
(
2014
).
56.
J.
Li
,
H.
Luo
,
L.
Wang
,
C.
Zhao
,
H.
Zhang
,
H.
Li
, and
Y.
Liu
,
Opt. Lett.
40
,
3659
(
2015
).
57.
P.
Yan
,
R.
Lin
,
S.
Ruan
,
A.
Liu
,
H.
Chen
,
Y.
Zheng
,
S.
Chen
,
C.
Guo
, and
J.
Hu
,
Sci. Rep.
5
,
8690
(
2015
).
58.
Y. I.
Jhon
,
J.
Lee
,
Y. M.
Jhon
, and
J. H.
Lee
,
IEEE J. Sel. Top. Quantum Electron.
24
,
1102208
(
2018
).
59.
L. C.
Kong
,
G. Q.
Xie
,
P.
Yuan
,
L. J.
Qian
,
S. X.
Wang
,
H. H.
Yu
, and
H. J.
Zhang
,
Photonics Res.
3
,
A47
(
2015
).
60.
Z.
Qin
,
G.
Xie
,
C.
Zhao
,
S.
Wen
,
P.
Yuan
, and
L.
Qian
,
Opt. Lett.
41
,
56
(
2016
).
61.
J.
Li
,
H.
Luo
,
B.
Zhai
,
R.
Lu
,
Z.
Guo
,
H.
Zhang
, and
Y.
Liu
,
Sci. Rep.
6
,
30361
(
2016
).
62.
Y.
Xie
,
L.
Kong
,
Z.
Qin
,
G.
Xie
, and
J.
Zhang
,
Opt. Eng.
55
,
081307
(
2016
).
63.
C.
Zhu
,
F.
Wang
,
Y.
Meng
,
X.
Yuan
,
F.
Xiu
,
H.
Luo
,
Y.
Wang
,
J.
Li
,
X.
Lv
,
L.
He
,
Y.
Xu
,
J.
Liu
,
C.
Zhang
,
Y.
Shi
,
R.
Zhang
, and
S.
Zhu
,
Nat. Commun.
8
,
14111
(
2017
).
64.
H.
Kataura
,
Y.
Kumazawa
,
Y.
Maniwa
,
I.
Umezu
,
S.
Suzuki
,
Y.
Ohtsuka
, and
Y.
Achiba
,
Synth. Met.
103
,
2555
(
1999
).
65.
P.
Chen
,
X.
Wu
,
X.
Sun
,
J.
Lin
,
W.
Ji
, and
K. L.
Tan
,
Phys. Rev. Lett.
82
,
2548
(
1999
).
66.
R. R.
Nair
,
P.
Blake
,
A. N.
Grigorenko
,
K. S.
Novoselov
,
T. J.
Booth
,
T.
Stauber
,
N. M. R.
Peres
, and
A. K.
Geim
,
Science
320
,
1308
(
2008
).
67.
Q.
Bao
,
H.
Zhang
,
Y.
Wang
,
Z.
Ni
,
Y.
Yan
,
Z. X.
Shen
,
K. P.
Loh
, and
D. Y.
Tang
,
Adv. Funct. Mater.
19
,
3077
(
2009
).
68.
A. H.
Castro Neto
,
F.
Guinea
,
N. M. R.
Peres
,
K. S.
Novoselov
, and
A. K.
Geim
,
Rev. Mod. Phys.
81
,
109
(
2009
).
69.
K. S.
Novoselov
,
A. K.
Geim
,
S. V.
Morozov
,
D.
Jiang
,
M. I.
Katsnelson
,
I. V.
Grigorieva
,
S. V.
Dubonos
, and
A. A.
Firsov
,
Nature
438
,
197
(
2005
).
70.
H.
Zhang
,
C.-X.
Liu
,
X.-L.
Qi
,
X.
Dai
,
Z.
Fang
, and
S.-C.
Zhang
,
Nat. Phys.
5
,
438
(
2009
).
71.
M. Z.
Hasan
and
C. L.
Kane
,
Rev. Mod. Phys.
82
,
3045
(
2010
).
72.
D.
Hsieh
,
D.
Qian
,
L.
Wray
,
Y.
Xia
,
Y. S.
Hor
,
R. J.
Cava
, and
M. Z.
Hasan
,
Nature
452
,
970
(
2008
).
73.
X.-L.
Qi
and
S.-C.
Zhang
,
Rev. Mod. Phys.
83
,
1057
(
2011
).
74.
S.
Lu
,
C.
Zhao
,
Y.
Zou
,
S.
Chen
,
Y.
Chen
,
Y.
Li
,
H.
Zhang
,
S.
Wen
, and
D.
Tang
,
Opt. Express
21
,
2072
(
2013
).
75.
M.
Hajlaoui
,
E.
Papalazarou
,
J.
Mauchain
,
G.
Lantz
,
N.
Moisan
,
D.
Boschetto
,
Z.
Jiang
,
I.
Miotkowski
,
Y. P.
Chen
,
A.
Taleb-Ibrahimi
,
L.
Perfetti
, and
M.
Marsi
,
Nano Lett.
12
,
3532
(
2012
).
76.
F.
Bernard
,
H.
Zhang
,
S. P.
Gorza
, and
P.
Emplit
, “
Towards mode-locked fiber laser using topological insulators
,” in
NP 2012, OSA Technical Digest
(Optical Society of America,
2012
), Paper No. NTh1A.5.
77.
S.
Chen
,
C.
Zhao
,
Y.
Li
,
H.
Huang
,
S.
Lu
,
H.
Zhang
, and
S.
Wen
,
Opt. Mater. Express
4
,
587
(
2014
).
78.
H.
Zhang
,
X.
He
,
W.
Lin
,
R.
Wei
,
F.
Zhang
,
X.
Du
,
G.
Dong
, and
J.
Qiu
,
Opt. Express
23
,
13376
(
2015
).
79.
J.
Lee
,
J.
Koo
,
J.
Lee
,
Y. M.
Jhon
, and
J. H.
Lee
,
Opt. Mater. Express
7
,
2968
(
2017
).
80.
J.
Wang
,
Z.
Jiang
,
H.
Chen
,
J.
Li
,
J.
Yin
,
J.
Wang
,
T.
He
,
P.
Yan
, and
S.
Ruan
,
Opt. Lett.
42
,
5010
(
2017
).
81.
J.
Wang
,
H.
Chen
,
Z.
Jiang
,
J.
Yin
,
J.
Wang
,
M.
Zhang
,
T.
He
,
J.
Li
,
P.
Yan
, and
S.
Ruan
,
Opt. Lett.
43
,
1998
(
2018
).
82.
J.
Wang
,
W.
Lu
,
J.
Li
,
H.
Chen
,
Z.
Jiang
,
J.
Wang
,
W.
Zhang
,
M.
Zhang
,
I. L.
Li
,
Z.
Xu
,
W.
Liu
, and
P.
Yan
,
IEEE J. Sel. Top. Quantum Electron.
24
,
1100706
(
2018
).
83.
T.
Zhen
,
W.
Kan
,
K.
Lingchen
,
Y.
Nan
,
W.
Yao
,
C.
Rong
,
H.
Weisheng
,
X.
Jianqiu
, and
T.
Yulong
,
Laser Phys. Lett.
12
,
065104
(
2015
).
84.
W.
Xiao-Fa
,
P.
Xiao-Ling
,
J.
Qiu-Xia
,
G.
Xiao-Hui
,
Z.
Jun-Hong
,
M.
Xue-Feng
, and
Y.
Su-Zhen
,
Chin. Phys. B
26
,
114205
(
2017
).
85.
L.
Cao
,
X.
Li
,
R.
Zhang
,
D.
Wu
,
S.
Dai
,
J.
Peng
,
J.
Weng
, and
Q.
Nie
,
Opt. Fiber Technol.
41
,
187
(
2018
).
86.
K. S.
Novoselov
,
D.
Jiang
,
F.
Schedin
,
T. J.
Booth
,
V. V.
Khotkevich
,
S. V.
Morozov
, and
A. K.
Geim
,
Proc. Natl. Acad. Sci. U. S. A.
102
,
10451
(
2005
).
87.
K. F.
Mak
,
C.
Lee
,
J.
Hone
,
J.
Shan
, and
T. F.
Heinz
,
Phys. Rev. Lett.
105
,
136805
(
2010
).
88.
Q. H.
Wang
,
K.
Kalantar-Zadeh
,
A.
Kis
,
J. N.
Coleman
, and
M. S.
Strano
,
Nat. Nanotechnol.
7
,
699
(
2012
).
89.
F.
Xia
,
H.
Wang
,
D.
Xiao
,
M.
Dubey
, and
A.
Ramasubramaniam
,
Nat. Photonics
8
,
899
(
2014
).
90.
K. F.
Mak
and
J.
Shan
,
Nat. Photonics
10
,
216
(
2016
).
91.
S.
Wang
,
H.
Yu
,
H.
Zhang
,
A.
Wang
,
M.
Zhao
,
Y.
Chen
,
L.
Mei
, and
J.
Wang
,
Adv. Mater.
26
,
3538
(
2014
).
92.
Y.
Wang
,
L.
Li
,
W.
Yao
,
S.
Song
,
J. T.
Sun
,
J.
Pan
,
X.
Ren
,
C.
Li
,
E.
Okunishi
,
Y.-Q.
Wang
,
E.
Wang
,
Y.
Shao
,
Y. Y.
Zhang
,
H.-T.
Yang
,
E. F.
Schwier
,
H.
Iwasawa
,
K.
Shimada
,
M.
Taniguchi
,
Z.
Cheng
,
S.
Zhou
,
S.
Du
,
S. J.
Pennycook
,
S. T.
Pantelides
, and
H.-J.
Gao
,
Nano Lett.
15
,
4013
(
2015
).
93.
X.
Yu
,
P.
Yu
,
D.
Wu
,
B.
Singh
,
Q.
Zeng
,
H.
Lin
,
W.
Zhou
,
J.
Lin
,
K.
Suenaga
,
Z.
Liu
, and
Q. J.
Wang
,
Nat. Commun.
9
,
1545
(
2018
).
94.
M.
Trushin
,
E. J. R.
Kelleher
, and
T.
Hasan
,
Phys. Rev. B
94
,
155301
(
2016
).
95.
H.
Liu
,
Y.
Du
,
Y.
Deng
, and
P. D.
Ye
,
Chem. Soc. Rev.
44
,
2732
(
2015
).
96.
L.
Li
,
Y.
Yu
,
G. J.
Ye
,
Q.
Ge
,
X.
Ou
,
H.
Wu
,
D.
Feng
,
X. H.
Chen
, and
Y.
Zhang
,
Nat. Nanotechnol.
9
,
372
(
2014
).
97.
S. P.
Koenig
,
R. A.
Doganov
,
H.
Schmidt
,
A. H.
Castro Neto
, and
B.
Özyilmaz
,
Appl. Phys. Lett.
104
,
103106
(
2014
).
98.
T.
Low
,
R.
Roldán
,
H.
Wang
,
F.
Xia
,
P.
Avouris
,
L. M.
Moreno
, and
F.
Guinea
,
Phys. Rev. Lett.
113
,
106802
(
2014
).
99.
F.
Xia
,
H.
Wang
, and
Y.
Jia
,
Nat. Commun.
5
,
4458
(
2014
).
100.
J.
Qiao
,
X.
Kong
,
Z.-X.
Hu
,
F.
Yang
, and
W.
Ji
,
Nat. Commun.
5
,
4475
(
2014
).
101.
S. B.
Lu
,
L. L.
Miao
,
Z. N.
Guo
,
X.
Qi
,
C. J.
Zhao
,
H.
Zhang
,
S. C.
Wen
,
D. Y.
Tang
, and
D. Y.
Fan
,
Opt. Express
23
,
11183
(
2015
).
102.
K.
Wang
,
B. M.
Szydłowska
,
G.
Wang
,
X.
Zhang
,
J. J.
Wang
,
J. J.
Magan
,
L.
Zhang
,
J. N.
Coleman
,
J.
Wang
, and
W. J.
Blau
,
ACS Nano
10
,
6923
(
2016
).
103.
J.
Herrmann
,
J. Opt. Soc. Am. B
11
,
498
(
1994
).
104.
D. E.
Spence
,
P. N.
Kean
, and
W.
Sibbett
,
Opt. Lett.
16
,
42
(
1991
).
105.
R. H.
Stolen
,
J.
Botineau
, and
A.
Ashkin
,
Opt. Lett.
7
,
512
(
1982
).
106.
M.
Hofer
,
M. E.
Fermann
,
F.
Haberl
,
M. H.
Ober
, and
A. J.
Schmidt
,
Opt. Lett.
16
,
502
(
1991
).
107.
N. J.
Doran
and
D.
Wood
,
Opt. Lett.
13
,
56
(
1988
).
108.
D. J.
Richardson
,
R. I.
Laming
,
D. N.
Payne
,
V.
Matsas
, and
M. W.
Phillips
,
Electron. Lett.
27
,
542
(
1991
).
109.
D. J.
Richardson
,
R. I.
Laming
,
D. N.
Payne
,
M. W.
Phillips
, and
V. J.
Matsas
,
Electron. Lett.
27
,
730
(
1991
).
110.
C. W.
Rudy
,
K. E.
Urbanek
,
M. J. F.
Digonnet
, and
R. L.
Byer
,
J. Lightwave Technol.
31
,
1809
(
2013
).
111.
J.
Li
,
Z.
Zhang
,
Z.
Sun
,
H.
Luo
,
Y.
Liu
,
Z.
Yan
,
C.
Mou
,
L.
Zhang
, and
S. K.
Turitsyn
,
Opt. Express
22
,
7875
(
2014
).
112.
L. F.
Johnson
,
J. Appl. Phys.
34
,
897
(
1963
).
113.
F.
Heine
,
E.
Heumann
,
G.
Huber
, and
K. L.
Schepler
,
Appl. Phys. Lett.
60
,
1161
(
1992
).
114.
J. F.
Pinto
,
L.
Esterowitz
, and
G. H.
Rosenblatt
,
Opt. Lett.
17
,
731
(
1992
).
115.
L. E.
Nelson
,
E. P.
Ippen
, and
H. A.
Haus
,
Appl. Phys. Lett.
67
,
19
(
1995
).
116.
R. C.
Sharp
,
D. E.
Spock
,
N.
Pan
, and
J.
Elliot
,
Opt. Lett.
21
,
881
(
1996
).
117.
K. L.
Schepler
,
B. D.
Smith
,
F.
Heine
, and
G.
Huber
,
Proc. SPIE
1864
,
186
(
1993
).
118.
G.
Galzerano
,
M.
Marano
,
S.
Longhi
,
E.
Sani
,
A.
Toncelli
,
M.
Tonelli
, and
P.
Laporta
,
Opt. Lett.
28
,
2085
(
2003
).
119.
D.
Gatti
,
G.
Galzerano
,
A.
Toncelli
,
M.
Tonelli
, and
P.
Laporta
,
Appl. Phys. B
86
,
269
(
2007
).
120.
K. L.
Schepler
,
B. D.
Smith
,
F.
Heine
, and
P. A.
Budni
, “
Mode-locking of diode-pumped Tm,Ho:YLF
,” in
ASSL 1994, OSA Technical Digest
(Opitcal Society of America,
1994
), Paper No. 257.
121.
E.
Sorokin
,
I. T.
Sorokina
,
A.
Unterhuber
,
E.
Wintner
,
A. I.
Zagumennyi
,
I. A.
Shcherbakov
,
V.
Carozza
,
A.
Toncelli
, and
M.
Tonelli
, “
A novel cw tunable and mode-locked 2 μm Cr,Tm,Ho:YSGG:GSAG laser
,” in
ASSL 1998, OSA Technical Digest
(Opitcal Society of America,
1998
), Paper No. 197.
122.
A. A.
Lagatsky
,
P.
Koopmann
,
O. L.
Antipov
,
C. T. A.
Brown
,
G.
Huber
, and
W.
Sibbett
, “
Femtosecond pulse generation with Tm-doped sesquioxides
,” in
CLEO 2013, OSA Technical Digest
(Optical Society of America,
2013
), Paper No. CA_6_3.
123.
Y.
Wang
,
W.
Jing
,
P.
Loiko
,
Y.
Zhao
,
H.
Huang
,
X.
Mateos
,
S.
Suomalainen
,
A.
Härkönen
,
M.
Guina
,
U.
Griebner
, and
V.
Petrov
,
Opt. Express
26
,
10299
(
2018
).
124.
H. A.
Haus
,
J. G.
Fujimoto
, and
E. P.
Ippen
,
J. Opt. Soc. Am. B
8
,
2068
(
1991
).
125.
J.
Ma
,
H. T.
Huang
,
K. J.
Ning
,
X. D.
Xu
,
G. Q.
Xie
,
L. J.
Qian
,
K. P.
Loh
, and
D. Y.
Tang
,
Opt. Lett.
41
,
890
(
2016
).
126.
J.
Ma
,
J.
Wang
,
D.
Shen
,
H.
Yu
,
H.
Zhang
, and
D.
Tang
,
Opt. Express
23
,
32311
(
2015
).
127.
C.
Hönninger
,
R.
Paschotta
,
M.
Graf
,
F.
Morier-Genoud
,
G.
Zhang
,
M.
Moser
,
S.
Biswal
,
J.
Nees
,
A.
Braun
,
G. A.
Mourou
,
I.
Johannsen
,
A.
Giesen
,
W.
Seeber
, and
U.
Keller
,
Appl. Phys. B
69
,
3
(
1999
).
128.
N.
Coluccelli
,
G.
Galzerano
,
D.
Gatti
,
A.
Di Lieto
,
M.
Tonelli
, and
P.
Laporta
,
Appl. Phys. B
101
,
75
(
2010
).
129.
J.
Ma
,
G. Q.
Xie
,
W. L.
Gao
,
P.
Yuan
,
L. J.
Qian
,
H. H.
Yu
,
H. J.
Zhang
, and
J. Y.
Wang
,
Opt. Lett.
37
,
1376
(
2012
).
130.
L. C.
Kong
,
Z. P.
Qin
,
G. Q.
Xie
,
X. D.
Xu
,
J.
Xu
,
P.
Yuan
, and
L. J.
Qian
,
Opt. Lett.
40
,
356
(
2015
).
131.
Z.
Qin
,
G.
Xie
,
L.
Kong
,
P.
Yuan
,
L.
Qian
,
X.
Xu
, and
J.
Xu
,
IEEE Photonics J.
7
,
1500205
(
2015
).
132.
Z.
Qiao
,
L.
Kong
,
G.
Xie
,
Z.
Qin
,
P.
Yuan
,
L.
Qian
,
X.
Xu
,
J.
Xu
, and
D.
Fan
,
Opt. Lett.
42
,
2547
(
2017
).
133.
L.
Kong
,
G.
Xie
,
Z.
Qin
,
X.
Xu
,
J.
Xu
,
P.
Yuan
, and
L.
Qian
, preprint arXiv:1707.03818 (
2017
).
134.
Z. P.
Qin
,
J. G.
Liu
,
G. Q.
Xie
,
J.
Ma
,
W. L.
Gao
,
L. J.
Qian
,
P.
Yuan
,
X. D.
Xu
,
J.
Xu
, and
D. H.
Zhou
,
Laser Phys.
23
,
105806
(
2013
).
135.
Y.
Wang
,
W.
Chen
,
M.
Mero
,
L.
Zhang
,
H.
Lin
,
Z.
Lin
,
G.
Zhang
,
F.
Rotermund
,
Y. J.
Cho
,
P.
Loiko
,
X.
Mateos
,
U.
Griebner
, and
V.
Petrov
,
Opt. Lett.
42
,
3076
(
2017
).
136.
J.
Zhang
,
K. F.
Mak
,
S.
Gröbmeyer
,
D.
Bauer
,
D.
Sutter
,
V.
Pervak
,
F.
Krausz
, and
O.
Pronin
, “
Generation of 220 fs, 20 W pulses at 2 μm from Kerr-lens mode-locked Ho:YAG thin-disk oscillator
,” in
CLEO/S&I 2017, OSA Technical Digest
(Optical Society of America,
2017
), Paper No. SM1I.6.
137.
M.
Tokurakawa
,
E.
Fujita
, and
C.
Kränkel
,
Opt. Lett.
42
,
3185
(
2017
).
138.
F.
Canbaz
,
I.
Yorulmaz
, and
A.
Sennaroglu
,
Opt. Lett.
42
,
3964
(
2017
).
139.
I. A.
Denisov
,
N. A.
Skoptsov
,
M. S.
Gaponenko
,
A. M.
Malyarevich
,
K. V.
Yumashev
, and
A. A.
Lipovskii
,
Opt. Lett.
34
,
3403
(
2009
).
140.
F.
Fusari
,
A. A.
Lagatsky
,
G.
Jose
,
S.
Calvez
,
A.
Jha
,
M. D.
Dawson
,
J. A.
Gupta
,
W.
Sibbett
, and
C. T. A.
Brown
,
Opt. Express
18
,
22090
(
2010
).
141.
M. S.
Gaponenko
,
V. E.
Kisel
,
N. V.
Kuleshov
,
A. M.
Malyarevich
,
K. V.
Yumashev
, and
A. A.
Onushchenko
,
Laser Phys. Lett.
7
,
286
(
2010
).
142.
K.
Yang
,
H.
Bromberger
,
H.
Ruf
,
H.
Schäfer
,
J.
Neuhaus
,
T.
Dekorsy
,
C. V.-B.
Grimm
,
M.
Helm
,
K.
Biermann
, and
H.
Künzel
,
Opt. Express
18
,
6537
(
2010
).
143.
N.
Coluccelli
,
A.
Lagatsky
,
A. D.
Lieto
,
M.
Tonelli
,
G.
Galzerano
,
W.
Sibbett
, and
P.
Laporta
,
Opt. Lett.
36
,
3209
(
2011
).
144.
A. A.
Lagatsky
and
J.-M.
Hopkins
, “
Diode-pumped femtosecond Tm-doped Lu2O3 ceramic laser
,” in
ASSL 2016, OSA Technical Digest
(Opitcal Society of America,
2016
), Paper No. JTu2A.5.
145.
K. J.
Yang
,
H.
Bromberger
,
D.
Heinecke
,
C.
Kölbl
,
H.
Schäfer
,
T.
Dekorsy
,
S. Z.
Zhao
,
L. H.
Zheng
,
J.
Xu
, and
G. J.
Zhao
,
Opt. Express
20
,
18630
(
2012
).
146.
B.
Yao
,
W.
Wang
,
K.
Yu
,
G.
Li
, and
Y.
Wang
,
Chin. Opt. Lett.
10
,
071402
(
2012
).
147.
K. J.
Yang
,
D. C.
Heinecke
,
C.
Kölbl
,
T.
Dekorsy
,
S. Z.
Zhao
,
L. H.
Zheng
,
J.
Xu
, and
G. J.
Zhao
,
Opt. Express
21
,
1574
(
2013
).
148.
H.
Cheng
,
X. D.
Jiang
,
X. P.
Hu
,
M. L.
Zhong
,
X. J.
Lv
, and
S. N.
Zhu
,
Opt. Lett.
39
,
2187
(
2014
).
149.
J.
Hou
,
B.
Zhang
,
X.
Su
,
R.
Zhao
,
Z.
Wang
,
F.
Lou
, and
J.
He
,
Opt. Commun.
347
,
88
(
2015
).
150.
T.
Feng
,
K.
Yang
,
J.
Zhao
,
S.
Zhao
,
W.
Qiao
,
T.
Li
,
T.
Dekorsy
,
J.
He
,
L.
Zheng
,
Q.
Wang
,
X.
Xu
,
L.
Su
, and
J.
Xu
,
Opt. Express
23
,
11819
(
2015
).
151.
Y.
Wang
,
R.
Lan
,
X.
Mateos
,
J.
Li
,
C.
Li
,
S.
Suomalainen
,
A.
Härkönen
,
M.
Guina
,
V.
Petrov
, and
U.
Griebner
,
Opt. Express
25
,
7084
(
2017
).
152.
W.
Zhou
,
X.
Fan
,
H.
Xue
,
R.
Xu
,
Y.
Zhao
,
X.
Xu
,
D.
Tang
, and
D.
Shen
,
Opt. Express
25
,
1815
(
2017
).
153.
Y.
Wang
,
R.
Lan
,
X.
Mateos
,
J.
Li
,
C.
Hu
,
C.
Li
,
S.
Suomalainen
,
A.
Härkönen
,
M.
Guina
,
V.
Petrov
, and
U.
Griebner
,
Opt. Express
24
,
18003
(
2016
).
154.
X.
Duan
,
J.
Yuan
,
Z.
Cui
,
B.
Yao
,
T.
Dai
,
J.
Li
, and
Y.
Pan
,
Appl. Opt.
55
,
1953
(
2016
).
155.
R.
Soulard
,
A.
Tyazhev
,
J.-L.
Doualan
,
A.
Braud
,
A.
Hideur
,
M.
Laroche
,
B.
Xu
, and
P.
Camy
,
Opt. Lett.
42
,
3534
(
2017
).
156.
F.
Haxsen
,
A.
Wienke
,
D.
Wandt
,
J.
Neumann
, and
D.
Kracht
,
Opt. Fiber Technol.
20
,
650
(
2014
).
157.
D. D.
Hudson
,
Opt. Fiber Technol.
20
,
631
(
2014
).
158.
C. W.
Rudy
,
M. J. F.
Digonnet
, and
R. L.
Byer
,
Opt. Fiber Technol.
20
,
642
(
2014
).
159.
I. T.
Sorokina
,
V. V.
Dvoyrin
,
N.
Tolstik
, and
E.
Sorokin
,
IEEE J. Sel. Top. Quantum Electron
20
,
0903412
(
2014
).
160.
S.
Kivisto
,
T.
Hakulinen
,
M.
Guina
, and
O. G.
Okhotnikov
,
IEEE Photonics Technol. Lett.
19
,
934
(
2007
).
161.
Q.
Wang
,
J.
Geng
,
T.
Luo
, and
S.
Jiang
,
Opt. Lett.
34
,
3616
(
2009
).
162.
S.
Kivisto
and
O. G.
Okhotnikov
,
IEEE Photonics Technol. Lett.
23
,
477
(
2011
).
163.
Q.
Wang
,
J. H.
Geng
,
Z.
Jiang
,
T.
Luo
, and
S. B.
Jiang
,
IEEE Photonics. Technol. Lett.
23
,
682
(
2011
).
164.
C. W.
Rudy
,
M. J. F.
Digonnet
,
R. L.
Byer
, and
S.
Jiang
, “
Thulium-doped germanosilicate mode-locked fiber lasers
,” in
FILAS 2012, OSA Technical Digest
(Optical Society of America,
2012
), Paper No. FTh4A.4.
165.
C.
Huang
,
C.
Wang
,
W.
Shang
,
N.
Yang
,
Y.
Tang
, and
J.
Xu
,
Sci. Rep.
5
,
13680
(
2015
).
166.
A. E.
Akosman
and
M. Y.
Sander
,
IEEE Photonics Technol. Lett.
28
,
1878
(
2016
).
167.
N.
Tolstik
,
E.
Sorokin
,
I.
Bugar
, and
I. T.
Sorokina
, “
Compact diode-pumped dispersion-managed SESAM-mode-locked Ho: Fiber laser
,” in
MICS 2016, OSA Technical Digest
(Optical Society of America,
2016
), Paper No. MM6C.4.
168.
W.
Yao
,
T.
Yulong
,
Y.
Shuo
, and
X.
Jianqiu
,
Laser Phys. Lett.
15
,
085101
(
2018
).
169.
M. A.
Solodyankin
,
E. D.
Obraztsova
,
A. S.
Lobach
,
A. I.
Chernov
,
A. V.
Tausenev
,
V. I.
Konov
, and
E. M.
Dianov
,
Opt. Lett.
33
,
1336
(
2008
).
170.
M. A.
Chernysheva
,
A. A.
Krylov
,
P. G.
Kryukov
,
N. R.
Arutyunyan
,
A. S.
Pozharov
,
E. D.
Obraztsova
, and
E. M.
Dianov
,
Opt. Express
20
,
B124
(
2012
).
171.
M. A.
Chernysheva
,
A. A.
Krylov
,
C.
Mou
,
R. N.
Arif
,
A. G.
Rozhin
,
M. H.
Rummelli
,
S. K.
Turitsyn
, and
E. M.
Dianov
,
IEEE J. Sel. Top. Quantum Electron.
20
,
1100908
(
2014
).
172.
N.
Li
,
M. Y.
Liu
,
X. J.
Gao
,
L.
Zhang
,
Z. X.
Jia
,
Y.
Feng
,
Y.
Ohishi
,
G. S.
Qin
, and
W. P.
Qin
,
Laser Phys. Lett.
13
,
075103
(
2016
).
173.
J.
Wang
,
X.
Liang
,
G.
Hu
,
Z.
Zheng
,
S.
Lin
,
D.
Ouyang
,
X.
Wu
,
P.
Yan
,
S.
Ruan
,
Z.
Sun
, and
T.
Hasan
,
Sci. Rep.
6
,
28885
(
2016
).
174.
Y.
Meng
,
Y.
Li
,
Y.
Xu
, and
F.
Wang
,
Sci. Rep.
7
,
45109
(
2017
).
175.
M.
Zhang
,
E. J. R.
Kelleher
,
F.
Torrisi
,
Z.
Sun
,
T.
Hasan
,
D.
Popa
,
F.
Wang
,
A. C.
Ferrari
,
S. V.
Popov
, and
J. R.
Taylor
,
Opt. Express
20
,
25077
(
2012
).
176.
Q.
Wang
,
T.
Chen
,
B.
Zhang
,
M.
Li
,
Y.
Lu
, and
K. P.
Chen
,
Appl. Phys. Lett.
102
,
131117
(
2013
).
177.
B.
Fu
,
Y.
Hua
,
X.
Xiao
,
H.
Zhu
,
Z.
Sun
, and
C.
Yang
,
IEEE J. Sel. Top. Quantum Electron.
20
,
1100705
(
2014
).
178.
M.
Jung
,
J.
Koo
,
J.
Park
,
Y.-W.
Song
,
K.
Lee
,
S. B.
Lee
, and
J. H.
Lee
, “
A femtosecond thulium holmium co-doped fiber based on graphene oxide evanescent filed interaction
,” in
CLEO/A&T 2014, OSA Technical Digest
(Optical Society of America,
2014
), Paper No. JW2A.20.
179.
V.
Dvoyrin
,
N.
Tolstik
,
E.
Sorokin
,
I.
Sorokina
, and
A.
Kurkov
, “
Graphene-mode-locked Holmium fiber laser operating beyond 2.1 μm
,” in
CLEO/Europe 2015, OSA Technical Digest
(Optical Society of America,
2015
), Paper No. CJ_7_4.
180.
G.
Sobon
,
Photonics Res.
3
,
A56
(
2015
).
181.
H.
Jeong
,
S. Y.
Choi
,
M. H.
Kim
,
F.
Rotermund
,
Y.-H.
Cha
,
D.-Y.
Jeong
,
S. B.
Lee
,
K.
Lee
, and
D.-I.
Yeom
,
Opt. Express
24
,
14152
(
2016
).
182.
J.
Sotor
,
M.
Pawliszewska
,
G.
Sobon
,
P.
Kaczmarek
,
A.
Przewolka
,
I.
Pasternak
,
J.
Cajzl
,
P.
Peterka
,
P.
Honzátko
,
I.
Kašík
,
W.
Strupinski
, and
K.
Abramski
,
Opt. Lett.
41
,
2592
(
2016
).
183.
J.
Sotor
,
J.
Bogusławski
,
T.
Martynkien
,
P.
Mergo
,
A.
Krajewska
,
A.
Przewłoka
,
W.
StrupiŃski
, and
G.
SoboŃ
,
Opt. Lett.
42
,
1592
(
2017
).
184.
M.
Pawliszewska
,
T.
Martynkien
,
A.
Przewłoka
, and
J.
Sotor
,
Opt. Lett.
43
,
38
(
2018
).
185.
M.
Jung
,
J.
Lee
,
J.
Koo
,
J.
Park
,
Y.-W.
Song
,
K.
Lee
,
S.
Lee
, and
J. H.
Lee
,
Opt. Express
22
,
7865
(
2014
).
186.
K.
Yin
,
B.
Zhang
,
L.
Li
,
T.
Jiang
,
X.
Zhou
, and
J.
Hou
,
Photonics Res.
3
,
72
(
2015
).
187.
M.
Jung
,
J.
Lee
,
W.
Song
,
Y. L.
Lee
,
J. H.
Lee
, and
W.
Shin
,
Laser Phys. Lett.
13
,
055103
(
2016
).
188.
J.
Tarka
,
J.
Boguslawski
,
R.
Zybala
,
M.
Kowalczyk
,
G.
Sobon
, and
J.
Sotor
,
Proc. SPIE
9728
,
972820
(
2016
).
189.
M.
Jung
,
J.
Lee
,
J.
Park
,
J.
Koo
,
Y. M.
Jhon
, and
J. H.
Lee
,
Opt. Express
23
,
19996
(
2015
).
190.
J.
Sotor
,
G.
Sobon
,
M.
Kowalczyk
,
W.
Macherzynski
,
P.
Paletko
, and
K. M.
Abramski
,
Opt. Lett.
40
,
3885
(
2015
).
191.
H.
Yu
,
X.
Zheng
,
K.
Yin
,
X. a
Cheng
, and
T.
Jiang
,
Appl. Opt.
54
,
10290
(
2015
).
192.
M.
Pawliszewska
,
Y.
Ge
,
Z.
Li
,
H.
Zhang
, and
J.
Sotor
,
Opt. Express
25
,
16916
(
2017
).
193.
Z.
Kang
,
M. Y.
Liu
,
X. J.
Gao
,
N.
Li
,
S. Y.
Yin
,
G. S.
Qin
, and
W. P.
Qin
,
Laser Phys. Lett.
12
,
045105
(
2015
).
194.
M. F.
Mohd Rusdi
,
A. A.
Latiff
,
M. C.
Paul
,
S.
Das
,
A.
Dhar
,
H.
Ahmad
, and
S. W.
Harun
,
Opt. Laser Technol.
89
,
16
(
2017
).
195.
H.
Ahmad
,
M. Z.
Samion
, and
N.
Yusoff
,
Opt. Fiber Technol.
45
,
122
(
2018
).
196.
J.
Jiang
,
C.
Mohr
,
J.
Bethge
,
A.
Mills
,
W.
Mefford
,
I.
Hartl
,
M. E.
Fermann
,
C.-C.
Lee
,
S.
Suzuki
,
T. R.
Schibli
,
N.
Leindecker
,
K. L.
Vodopyanov
, and
P. G.
Schunemann
, “
500 MHz, 58 fs highly coherent Tm fiber soliton laser
,” in
CLEO/S&I 2012, OSA Technical Digest
(Optical Society of America,
2012
), Paper No. CTh5D.7.
197.
F.
Haxsen
,
A.
Ruehl
,
M.
Engelbrecht
,
D.
Wandt
,
U.
Morgner
, and
D.
Kracht
,
Opt. Express
16
,
20471
(
2008
).
198.
A.
Wienke
,
F.
Haxsen
,
D.
Wandt
,
U.
Morgner
,
J.
Neumann
, and
D.
Kracht
,
Opt. Lett.
37
,
2466
(
2012
).
199.
Y.
Nomura
and
T.
Fuji
,
Opt. Express
22
,
12461
(
2014
).
200.
Y.
Tang
,
A.
Chong
, and
F. W.
Wise
,
Opt. Lett.
40
,
2361
(
2015
).
201.
B.
Sun
,
J.
Luo
,
B. P.
Ng
, and
X.
Yu
,
Opt. Lett.
41
,
4052
(
2016
).
202.
B.
Sun
,
J.
Luo
,
Z.
Yan
,
K.
Liu
,
J.
Ji
,
Y.
Zhang
,
Q. J.
Wang
, and
X.
Yu
,
Opt. Express
25
,
8997
(
2017
).
203.
M.
Engelbrecht
,
F.
Haxsen
,
A.
Ruehl
,
D.
Wandt
, and
D.
Kracht
,
Opt. Lett.
33
,
690
(
2008
).
204.
P.
Li
,
A.
Ruehl
,
U.
Grosse-Wortmann
, and
I.
Hartl
,
Opt. Lett.
39
,
6859
(
2014
).
205.
D. I. M.
Zen
,
N.
Saidin
,
S. S. A.
Damanhuri
,
S. W.
Harun
,
H.
Ahmad
,
M. A.
Ismail
,
K.
Dimyati
,
A.
Halder
,
M. C.
Paul
,
S.
Das
,
M.
Pal
, and
S. K.
Bhadra
,
Appl. Opt.
52
,
1226
(
2013
).
206.
J.
Liu
,
S.
Wu
,
J.
Xu
,
Q.
Wang
,
Q.-H.
Yang
, and
P.
Wang
, “
Mode-locked 2 μm Thulium-doped fiber laser with graphene oxide saturable absorber
,” in
CLEO/S&I, OSA Technical Digest
(Optical Society of America,
2012
), Paper No. JW2A.76.
207.
D. A.
Gaponov
,
R.
Dauliat
,
D.
Darwich
,
T.
Mansuryan
,
R.
Jamier
,
S.
Grimm
,
K.
Schuster
, and
P.
Roy
,
J. Opt. Soc. Am. B
32
,
1656
(
2015
).
208.
L. D.
DeLoach
,
R. H.
Page
,
G. D.
Wilke
,
S. A.
Payne
, and
W. F.
Krupke
,
IEEE J. Quantum Electron.
32
,
885
(
1996
).
209.
S. B.
Mirov
,
V. V.
Fedorov
,
D.
Martyshkin
,
I. S.
Moskalev
,
M.
Mirov
, and
S.
Vasilyev
,
IEEE J. Sel. Top. Quantum Electron.
21
,
292
(
2015
).
210.
U.
Demirbas
and
A.
Sennaroglu
,
Opt. Lett.
31
,
2293
(
2006
).
211.
E.
Sorokin
,
I. T.
Sorokina
,
M. S.
Mirov
,
V. V.
Fedorov
,
I. S.
Moskalev
, and
S. B.
Mirov
, “
Ultrabroad continuous-wave tuning of ceramic Cr:ZnSe and Cr:ZnS lasers
,” in
ASSL 2010, OSA Technical Digest
(Optical Society of America,
2010
), Paper No. AMC2.
212.
I.
Moskalev
,
S.
Mirov
,
M.
Mirov
,
S.
Vasilyev
,
V.
Smolski
,
A.
Zakrevskiy
, and
V.
Gapontsev
,
Opt. Express
24
,
21090
(
2016
).
213.
E.
Sorokin
,
S.
Naumov
, and
I. T.
Sorokina
,
IEEE J. Sel. Top. Quantum Electron.
11
,
690
(
2005
).
214.
T. J.
Carrig
,
G. J.
Wagner
,
A.
Sennaroglu
,
J. Y.
Jeong
, and
C. R.
Pollock
, “
Acousto-optic mode-locking of a Cr2+:ZnSe laser
,” in
ASSL 2000, OSA Technical Digest
(Optical Society of America,
2000
), Paper No. 182.
215.
T. J.
Carrig
,
G. J.
Wagner
,
A.
Sennaroglu
,
J. Y.
Jeong
, and
C. R.
Pollock
,
Opt. Lett.
25
,
168
(
2000
).
216.
I. T.
Sorokina
,
E.
Sorokin
,
A. D.
Lieto
,
M.
Tonelli
,
R. H.
Page
, and
K. I.
Schaffers
, “
Active and passive mode-locking of the Cr2+:ZnSe laser
,” in
ASSL 2001, OSA Technical Digest
(Optical Society of America,
2001
), Paper No. MC2.
217.
E.
Sorokin
,
I. T.
Sorokina
,
A. D.
Lieto
,
M.
Tonelli
, and
P.
Minguzzi
, “
Mode-locked ceramic Cr2+:ZnSe laser
,” in
ASSL 2003, OSA Technical Digest
(Optical Society of America,
2003
), Paper No. 227.
218.
C. R.
Pollock
,
N. A.
Brilliant
,
D.
Gwin
,
T. J.
Carrig
,
W. J.
Alford
,
J. B.
Heroux
,
W. I.
Wang
,
I.
Vurgaftman
, and
J. R.
Meyer
, “
Mode locked and Q-switched Cr:ZnSe laser using a semiconductor saturable absorbing mirror (SESAM)
,” in
ASSL 2005, OSA Technical Digest
(Optical Society of America,
2005
), Paper No. TuA6.
219.
I. T.
Sorokina
,
E.
Sorokin
,
T. J.
Carrig
, and
K. I.
Schaffers
, “
A SESAM passively mode-locked Cr:ZnS laser
,” in
ASSL 2006, OSA Technical Digest
(Optical Society of America,
2006
), Paper No. TuA4.
220.
I. T.
Sorokina
,
E.
Sorokin
, and
T. J.
Carrig
, “
Femtosecond pulse generation from a SESAM mode-locked Cr:ZnSe laser
,” in
CLEO/S&I 2006, OSA Techincal Digest
(Optical Society of America,
2006
), Paper No. CMQ2.
221.
I. T.
Sorokina
and
E.
Sorokin
, “
Chirped-mirror dispersion controlled femtosecond Cr:ZnSe laser
,” in
ASSL 2007, OSA Technical Digest
(Optical Society of America,
2007
), Paper No. WA7.
222.
E.
Sorokin
and
I. T.
Sorokina
, “
Femtosecond operation and random quasi-phase-matched self-doubling of ceramic Cr:ZnSe laser
,” in
CLEO/QELS 2010, OSA Technical Digest
(Optical Society of America,
2010
), Paper No. CTuGG2.
223.
E.
Sorokin
,
N.
Tolstik
, and
I. T.
Sorokina
, “
Femtosecond operation and self-doubling of Cr:ZnS laser
,” in
NLO 2011, OSA Technical Digest
(Optical Society of America,
2011
), Paper No. NThC1.
224.
E.
Sorokin
,
N.
Tolstik
,
K. I.
Schaffers
, and
I. T.
Sorokina
,
Opt. Express
20
,
28947
(
2012
).
225.
M. N.
Cizmeciyan
,
H.
Cankaya
,
A.
Kurt
, and
A.
Sennaroglu
,
Opt. Lett.
34
,
3056
(
2009
).
226.
E.
Sorokin
and
I. T.
Sorokina
, “
Ultrashort-pulsed Kerr-lens modelocked Cr:ZnSe laser
,” in
CLEO/Europe 2009, OSA Technical Digest
(Optical Society of America,
2009
), Paper No. CF1_3.
227.
M. N.
Cizmeciyan
,
H.
Cankaya
,
A.
Kurt
, and
A.
Sennaroglu
,
Appl. Phys. B
106
,
887
(
2012
).
228.
N.
Tolstik
,
E.
Sorokin
, and
I. T.
Sorokina
,
Opt. Lett.
38
,
299
(
2013
).
229.
N.
Tolstik
,
E.
Sorokin
, and
I. T.
Sorokina
, “
Watt-level Kerr-Lens mode-locked Cr:ZnS laser at 2.4 μm
,” in
CLEO/S&I 2013, OSA Technical Digest
(Optical Society of America,
2013
), Paper No. CTh1H.2.
230.
E.
Sorokin
,
N.
Tolstik
, and
I. T.
Sorokina
,
Proc. SPIE
8599
,
859916
(
2013
).
231.
S.
Vasilyev
,
M.
Mirov
, and
V.
Gapontsev
, “
High power Kerr-Lens mode-locked femtosecond mid-IR laser with efficient second harmonic generation in polycrystalline Cr2+:ZnS and Cr2+:ZnSe
,” in
ASSL 2014, OSA Technical Digest
(Optical Society of America,
2014
), Paper No. AM3A.3.
232.
S.
Vasilyev
,
I.
Moskalev
,
M.
Mirov
,
V.
Smolski
,
S.
Mirov
, and
V.
Gapontsev
,
Proc. SPIE
9731
,
97310B
(
2016
).
233.
S.
Vasilyev
,
M.
Mirov
, and
V.
Gapontsev
,
Opt. Express
22
,
5118
(
2014
).
234.
S.
Vasilyev
,
M.
Mirov
, and
V.
Gapontsev
, “
Mid-IR Kerr-Lens mode-locked polycrystalline Cr2+:ZnS laser with 0.5 MW peak power
,” in
ASSL 2015, OSA Technical Digest
(Optical Society of America,
2015
), Paper No. AW4A.3.
235.
S.
Vasilyev
,
I.
Moskalev
,
M.
Mirov
,
V.
Smolski
,
S.
Mirov
, and
V.
Gapontsev
, “
Kerr-Lens mode-locked middle IR polycrystalline Cr:ZnS laser with a repetition rate 1.2 GHz
,” in
ASSL 2016, OSA Technical Digest
(Optical Society of America,
2016
), Paper No. AW1A.2.
236.
J. B.
Dherbecourt
,
A.
Denoeud
,
J. M.
Melkonian
,
M.
Raybaut
,
A.
Godard
,
M.
Lefebvre
, and
E.
Rosencher
,
Opt. Lett.
36
,
751
(
2011
).
237.
C.
Hu
,
J.
Zhu
,
Z.
Wang
,
X.
Sun
,
L.
Wei
,
W.
Ling
,
Z.
Dong
,
Q.
Wang
,
C.
Gao
, and
Z.
Wei
,
Chin. Phys. B
26
,
014206
(
2017
).
238.
M.
Fan
,
T.
Li
,
G.
Li
,
H.
Ma
,
S.
Zhao
,
K.
Yang
, and
C.
Kränkel
,
Opt. Lett.
42
,
286
(
2017
).
239.
W.
Ma
,
X.
Qian
,
J.
Wang
,
J.
Liu
,
X.
Fan
,
J.
Liu
,
L.
Su
, and
J.
Xu
,
Sci. Rep.
6
,
36635
(
2016
).
240.
Y. O.
Aydın
,
V.
Fortin
,
F.
Maes
,
F.
Jobin
,
S. D.
Jackson
,
R.
Vallée
, and
M.
Bernier
,
Optica
4
,
235
(
2017
).
241.
Y. O.
Aydin
,
V.
Fortin
,
R.
Vallée
, and
M.
Bernier
,
Opt. Lett.
43
,
4542
(
2018
).
242.
F.
Maes
,
V.
Fortin
,
S.
Poulain
,
M.
Poulain
,
J.-Y.
Carrée
,
M.
Bernier
, and
R.
Vallée
,
Optica
5
,
761
(
2018
).
243.
P.
Paradis
,
V.
Fortin
,
Y. O.
Aydin
,
R.
Vallée
, and
M.
Bernier
,
Opt. Lett.
43
,
3196
(
2018
).
244.
Z.
Qin
,
G.
Xie
,
J.
Ma
,
P.
Yuan
, and
L.
Qian
,
Chin. Opt. Lett.
15
,
111402
(
2017
).
245.
S.
Tokita
,
M.
Murakami
,
S.
Shimizu
,
M.
Hashida
, and
S.
Sakabe
,
Opt. Lett.
36
,
2812
(
2011
).
246.
C.
Frerichs
and
U. B.
Unrau
,
Opt. Fiber Technol.
2
,
358
(
1996
).
247.
J.
Li
,
D. D.
Hudson
,
Y.
Liu
, and
S. D.
Jackson
,
Opt. Lett.
37
,
3747
(
2012
).
248.
C.
Wei
,
X.
Zhu
,
R. A.
Norwood
, and
N.
Peyghambarian
,
Opt. Lett.
37
,
3849
(
2012
).
249.
A.
Haboucha
,
V.
Fortin
,
M.
Bernier
,
J.
Genest
,
Y.
Messaddeq
, and
R.
Vallée
,
Opt. Lett.
39
,
3294
(
2014
).
250.
T.
Hu
,
D. D.
Hudson
, and
S. D.
Jackson
,
Opt. Lett.
39
,
2133
(
2014
).
251.
P.
Tang
,
Z.
Qin
,
J.
Liu
,