Recent advances in laser diode-pumped InnoSlab amplifiers

Abstract

Significant progress has been made in high-power ultrafast laser technology since the development of diode-pumped solid-state laser systems. Three main types of diode-pumped laser systems, InnoSlab, fiber, and thin disk lasers, offer highly efficient cooling geometries that are essential for high-power ultrafast amplifiers. These systems employ amplifier chain configurations customized to their individual geometries, scaling the low-power seed lasers to high power via multi-pass, multi-stage, and regenerative amplification techniques. The partially end-pumped InnoSlab amplifier is distinguished by its slab-shaped gain medium and a highly compact design. This design offers a large surface-to-volume ratio, moderate gain per pass, and reduced nonlinear effects, facilitating the amplification of low-power ultrafast seed laser pulses to kilowatt-level output power at high repetition rates in the multi-MHz range. This review highlights the characteristics of InnoSlab technology and its amplifier configurations, discussing recent advancements in new cavity designs aimed at enhancing gain and beam quality. Additionally, it covers the mechanisms of generating high peak power few-cycle pulses, including non-linear post-pulse compression. The review also explores the potential applications of InnoSlab systems for generating extreme ultraviolet (XUV) and terahertz (THz) frequencies.

Graphical abstract

Keywords

ultrafast lasers / InnoSlab amplifier / multi-pass cell / hybrid resonator / Yb-doped materials

1 Introduction

High-power ultrashort laser technology has become an indispensable tool in industrial and scientific applications. The growing need for important applications of ultrafast lasers is constantly pushing the development of novel laser technologies with high average power, peak power, and repetition rate. Solid-state lasers dominate the current market, but they face challenges of pronounced thermal effects at high pumping power. These effects can hinder the scaling of average power while maintaining high beam quality. For efficient power scaling, a laser gain medium design with a high aspect ratio is typically beneficial, as it facilitates better heat removal and supports kilowatt-level power at high efficiency and beam quality. The emergence of laser diodes (LDs) [1] has resulted in the development of three main diode-pumped solid-state lasers (DPSSLs) configurations: fiber, thin-disk (TD), and InnoSlab lasers. These laser designs differ in gain medium geometry, pumping methods, and amplification setups. In continuous wave (CW) laser operation, benefiting from the large surface area for efficient heat dissipation, the average output power of >100 kW [2] in fiber lasers, ≥ 24 kW [3] in thin-disk, and > 5 kW [4] with InnoSlab has been reported. For the generation and amplification of ultrafast laser pulses, these three laser architectures have achieved high-average powers ranging from several watts to kilowatts, with peak powers reaching the megawatt to terawatt range. The fiber lasers, characterized by their low mode area and long amplification length, provide high single-pass gain among the three geometries. The excellent heat removal capability due to a large surface-to-volume ratio and no thermal lensing enables the high beam quality operation. However, the large accumulation of non-linearities per pass in fiber laser due to long interaction length and the low energy storage capacity complicates the high-power ultrafast operation in fs-regime. Similarly, the self-focusing threshold of glass limits the further increase of peak power [5]. The single fiber channel can not provide high peak power intense pulses for dominance of non-linearity, and amplification configurations for higher energies are based on chirped pulse amplification CPA, coherent beam combination CBC, and divide beam Combination DPA [6, 7]. The thin disk laser concept [8] is the disk-shaped gain medium, typically with a thickness of 100 to 300 μm and a much larger diameter of up to several mm (diameter $\gg$ thickness). The short interaction length of ~100 μm (TD-thickness), higher energy storage capacity than fiber, and low nonlinear effects facilitate the ultrafast operation at the kilowatt level. The TD geometry offers significant advantages for ultrafast operation via regenerative and multi-pass amplification [9, 10]. The regenerative amplifiers are typically limited to repetition rates of 1–2 MHz set by switching voltage necessary to operate the Pockels cell (PC); it also limits the pulse energy scaling due to high nonlinearities arsis in PC [11]. On the other hand, a thin-disk multipass amplifier TDMPA is an important configuration for scaling the average output power to a multi-KW level and pulse energies to a multi-100 mJ level at ultrafast pulse duration without requiring CPA and CBC.

The InnoSlab amplifiers are classified as a moderate approach to ultrafast power amplification. They are positioned between fiber and TD amplifiers due to their moderate gain per pass and low non-linear effects. This laser system is characterized by its rectangular slab-shaped design and hybrid resonator. The geometry is based on neodymium and ytterbium-doped materials, which are efficiently pumped by InGaAsP/GaAs or AlGaAs/Ga/As LDs at 808−888 nm and 900−980 nm, respectively [12]. Notably, the partially end-pumped InnoSlab geometry utilizes high-power LD pump sources more efficiently than the other two geometries, which require more or less complicated pumping schemes. Initially, it relied on a four-level Nd-doped material for their lower pump saturation intensities (~2 times lower than ytterbium-doped materials), which typically requires nearly zero minimum transparency pump intensity (I_min ≈ 0) [13]. Whereas in ytterbium-doped materials, pumping with a high brightness LD source rather than at high power is essential to achieve minimal pump intensities ($I_{\mathrm{m}\mathrm{i}\mathrm{n}}=2.8\mathrm{k}\mathrm{W}/{\mathrm{c}\mathrm{m}}^2$, for Yb:YAG) while avoiding the thermal effects of high pumping power. This is because pumping with low brightness LD requires much higher power to achieve desired intensities. The emergence of high-brightness LDs enables the use of ytterbium-doped materials with InnoSlab in continuous and high-power ultrafast regimes with significantly improved efficiency, reliability, and compactness. Apart from laser operation at 1 μm regime, the InnoSlab concept is well suited for operation at 2.0 μm ranges with high power. In the last two decades, the InnoSlab architecture has boosted few-watt picosecond and femtosecond mode-coupled laser sources to kW level [14, 15]. Furthermore, ns-pulses up to 500 mJ of pulse energies have also been realized efficiently [16–18].

The three thin-disk, fiber, and InnoSlab amplifiers are comparable in terms of power scaling and efficiency, with the first two sitting at upper and lower extremes in terms of gain and non-linearities. The fiber laser provides a high gain per pass and, at the same time, very high nonlinearities. On the other hand, a thin disk laser, despite offering a small gain per pass, takes advantage of power scaling at an ultrafast regime due to very small nonlinearities. Meanwhile, with moderate gain and moderate non-linearities, the InnoSlab amplifier is a balanced technology and is placed in between these technologies. The review focuses on the significant advancements in high-power ultrafast InnoSlab lasers over the past decade, offering insights into various operational regimes. It highlights the advantages of its design and recent developments in cavity configurations that enhance amplifier performance. The review also describes the principle techniques employed in InnoSlab to generate high-power ultrafast pulses beyond the spectral bandwidth limit of the gain medium. Additionally, it evaluates the performances of InnoSlab amplifiers for XUV and THz radiation generation and discusses the key characteristics of Yb-doped materials essential for high-power ultrafast lasers.

2 InnoSlab laser technology

2.1 Design consideration

The InnoSlab technology was developed by Fraunhofer ILT 28 years ago, in 1996 [19]. Soon after its invention, laser setups with hybrid resonators were demonstrated with this concept for CW and ultrafast pulse regime [20, 21]; for the evaluation of InnoSlab technology, see Fig.1(I)−(III). The core component of this technology is a six-faced rectangular-shaped slab crystal, Fig.1(IV). Compared to traditional slab designs, the laser beam takes a straight path along the laser axis in $z\text{-}\mathrm{a}\mathrm{x}\mathrm{i}\mathrm{s}$ instead of a zigzag; thus, only two optical end-sides of the slab ($a\times c$) are coated for laser and pump radiations, making it a cost-effective design. The two sides ($b\times c$) remain free, while two large surfaces (a × b) bring the advent of effective cooling through thermal contact with water-cooled heat sinks. Water-cooled heat sinks omit the direct contact of crystal with the cooling water and reduce the acoustic noise by the water flow to the amplified signal. With a unidirectional heat flow, a thermal gradient appears in the direction of heat flow, causing uniform thermal lensing in the vertical direction. InnoSlab is partially end-pumped on a rectangular cross-section of a high aspect ratio ($a\times c$) by a stack of LDs. In end-pumping, the radiations are collinear to laser mode and can be more efficiently absorbed than transverse pumping [1]. The slab geometry perfectly matches the characteristics of laser diode bars, and the terms slow and fast-axis in LD are consistent with the $x$- and $y$-direction of the slab, respectively. For pumping, the LD pump beam is transformed into a thin homogenized rectangular sheet with a top-hat intensity profile along the slow axis and a Gaussian distribution profile along the fast axis via a pump-shaping module. The pumping sheet fills the whole slab width in the horizontal direction and partially fills the vertical direction in the middle. The pump beam shaping module consists of slow and fast axis focusing and collimating cylindrical lenses, a slow axis mixing duct/waveguide to homogenize the pump beam, and spherical lenses to turn the pump beam into a homogeneous rectangular shape. Homogeneous pumping is critical to mitigate the thermal distortion along the slow axis. Fig.1(V) and (VI) illustrate the pumping mechanism of the InnoSlab laser. Generally, the transmission efficiency of the imaging system is > 90%. It can be better tuned using fine optics and high-brightness LDs. Typically, two resonator configurations were adopted for power extraction: (i) stable resonator and (ii) hybrid resonator. An earlier development adopted a stable resonator providing an elliptical beam profile in a high aspect ratio with high beam quality in the fast axis and high order mode in the slow axis [22]. Later, a simple two-mirror stable-unstable hybrid resonator is applied to operate the InnoSlab with symmetrical beam quality and optimal efficiency [20, 22]. For the hybrid resonator, the corresponding output coupling (OC) can be commuted as $T=1-1/M$, where $M=R_{\mathrm{i}\mathrm{n}}/R_{\mathrm{o}\mathrm{u}\mathrm{t}}$ is the magnification factor of the input and output coupler.

Fig.1 The evolution of InnoSlab technology: (I) patent of partially-end pumped slab laser filed in 1997 [19]; (II) pump mechanism for InnoSlab first introduced in 1998 [20]; (III) patent of partially end pumped InnoSlab amplifier filed in 2000 [21]; (IV) slab crystal dimension; (V) pump beam shaping system and intensity distribution inside slab crystal along fast axis; (VI) partially end pumped InnoSlab laser with hybrid cavity.

Full size|PPT slide

2.2 InnoSlab amplifier design

In the power amplifier configuration, the InnoSlab amplifier patented in 1998 [21] is integrated to amplify a low-power seed laser oscillator, keeping the other laser beam parameters intact. The main amplifier consists of a partially end-pumped slab amplifier and a hybrid resonator (Fig.2). The pumping scheme of the amplifier is the same as discussed before. A vertically mode-matched seed beam is injected into the slab amplifier, which takes multiple passes through the crystal between resonator mirrors. The resonator remains stable along the fast axis as a uniform thermal lens reproduces the laser mode in every pass. In contrast, the beam size increases in the slow axis in each pass, depending upon the beam divergence and cavity mirror magnification factor, which also decides the number of passes through a given slab. Different hybrid resonator configurations have been adopted with InnoSlab; a detailed discussion is given in the next section. The increasingly large beam size inside the gain medium in an unstable direction brings multiple advantages simultaneously; it increases efficiency due to constant gain saturation and keeps the intensity away from the damage threshold. However, in order to keep the amplifying beam intensity sufficient for gain saturation, one can not take any arbitrary value of the magnification factor and divergence angle for beam expansion. The constant seed laser to saturation intensity ratio and short interaction length further keep the non-linearities (B-integral) low and enable the generation of high average power bandwidth-limited pulses. This also enables power scalability without deteriorating thermal management, efficiency, and beam quality [23]. Thus, InnoSlab can increase output power from several hundred watts to kilowatt levels in ultrafast pulsed operation. For efficient power scaling at high beam quality, mode matching of the seed laser to that of the amplifier is critical. Along the fast axis, mode matching is always excellent since the cavity remains stable for a broad operation range due to uniform thermal lens. Similarly, through careful optimization of seed beam intensity, diameter, and tight overlap between consecutive passes, i.e., to keep the separation between two consecutive passes as minimum as possible, it is possible to use maximal gain volume efficiently in the slow axis. It is important to note that although the beam passes multiple times through gain volumes, in contrast to TD-multipass, it is only a single-pass amplifier since, in each roundtrip, a new section of the pumped gain volume is saturated, minimizing the effect of aberrations. A typical InnoSlab amplifier offers a moderate gain of 2–10 per pass and supports a large amplification factor of up to 1000 [24]. A more detailed discussion about InnoSlab amplifier design and its advantages is given in Ref. [23].

Fig.2 Configuration of partially double-end pumped InnoSlab amplifier.

Full size|PPT slide

Recently, a new innovative slab amplifier design was introduced [25] following the InnoSlab configuration. This design expands the InnoSlab concept by incorporating zigzag geometry to enhance beam quality and increase the gain per pass. The author has termed this new approach “Innozag”, merging the principles of both InnoSlab and zigzag slab geometry. In the Innozag design, both the pump and laser beams follow a zigzag pathway through total internal reflection (TIR), as opposed to the straightforward route of the InnoSlab. This allows for extended seed beam interaction within the gain medium. At TIR points, intensities for both pump and laser beams increase two-fold due to overlapping, thereby significantly enhancing the gain in these regions. In proof of studies, a 13% increase in amplification factor was observed as compared to typical InnoSlab under the same condition. However, for TIR both large surfaces must be polished and coated with high reflectivity (HR) coatings for the pump and seed beams, which were not coated in the traditional InnoSlab setup. Section 2.4 will elaborate on beam quality improvements using this concept.

2.3 Cavity schemes for InnoSlab amplifiers

In the InnoSlab amplifier, various hybrid cavity configurations have been integrated to realize power amplification. Traditionally, a confocal arrangement of two cylindrical mirrors (concave−convex: positive branch off-axis confocal cavity; concave−concave: negative-branch confocal cavity) was introduced in the InnoSlab amplifier arrangement, as shown in Fig.3(a). The distance between two cavity mirrors together with the thermal lens determines the mode matching along the fast axis. In large traverse dimensions, the confocal arrangement forms a hybrid resonator for beam expansion. Nevertheless, the confocal arrangement has limitations like fixed magnification and requires careful alignment for beam divergence. Likewise, a plane−plane mirror hybrid cavity is also applied to conveniently adjust amplifying beam divergence along the slow axis, as the beam divergence remains constant [26]. It consisted of two plane mirrors, where OC is wedged to fold the beam multiple times with inherited beam divergence, as shown in Fig.3(b). Prior to being injected into the amplifier portion, the seed beam waist position and divergence are accurately adjusted. Therefore, by manipulating the parameters of the injected seed beam, magnification in each pass is only a result of the inherent beam divergence. However, the main challenge in this configuration is to keep the wedge angle as small as possible while avoiding amplified spontaneous emission ASE and to utiliz the pumped gain area effectively. Recently, another plane−convex hybrid cavity was introduced for efficient mode matching and suppressing self-lasing [27]. This cavity configuration is more compact and offers better efficiency. In addition to the two mirror hybrid cavities, the InnoSlab amplifier has also incorporated another multi-mirror cavity layout. The multiple mirror cavities [28] involve a series of reflective mirrors to redirect the beam for many passes through the slab amplifier in a discrete path (DP) instead of using a single OC [Fig.3(c) and (d)]. This arrangement enables the independent manipulation of each pass through the slab by placing one mirror to individual control one pass, in contrast to conventional two-mirror cavities. The DP configuration offers various advantages: (i) independent adjustment of corresponding mirrors and hence the beam path, (ii) reduces the fitting sensitivity of reflective mirrors, (iii) the magnification factor for beam expansion can independently be adjusted for each pass according to $M=R_{\mathrm{i}\mathrm{n}}/R_k(k=1,2,3,\dots )$, R_k output mirror with a given radius of curvature (iv) and significantly eliminates self-lasing and suppresses ASE because all cavity mirrors are not set parallel to each other [28, 29].

Fig.3 Different cavity configurations for single end-pumped InnoSlab amplifiers: (a) off-axis positive-branch confocal unstable hybrid resonator [30], (b) plane-plane mirror hybrid resonator [26], (c) multi mirror with discrete beam path hybrid resonator [31], and (d) ray tracing with DP InnoSlab amplifier [32].

Full size|PPT slide

2.4 Thermal effects

An efficient heat removal scheme is essential to support high pump intensities for high-power laser operation at high beam quality. Thermally-induced aberrations in laser radiation are the main challenge in limiting beam quality. These aberrations are characterized by phase difference, which refers to the phase difference between the centre and the edge of the laser beam acquired at one pass through the crystal. As the beam quality is an important parameter in characterizing the laser output, and any passive optical transformations can not transform it. The beams close to the diffraction limit are beneficial in focusing the beam with the desired spot size. For good beam quality, thermal load management is vital to minimize the aberrations of the thermal lens. In the slab-shaped gain medium, achieving high-beam quality amplified pulses is particularly challenging since the seed beam has to pass many times without a waveguide, and an asymmetric thermal lens exists due to the asymmetric structure of the gain medium [33]. An optimized gain geometry efficiently reduces thermal load by minimizing the distance that radiation travels to reach the heat sink while providing a larger cooling surface area. The InnoSlab design, featuring larger surfaces directly attached to the heat sink, facilitates effective heat removal. By increasing the length $b$ of the slab, the contact area with the heat sink can be expanded without inducing amplified spontaneous emission (ASE) effects, although this may lead to an accumulation of more nonlinearities. The one-dimensional heat flow in the y-direction establishes a homogeneous cylindrical thermal lens and avoids depolarization by birefringence. For a given pump power, the focal length of the thermal lens in the y-direction can be very small $f_{th}=k{S}_{ac}/(\chi .{\eta }_h{P}_{ab})$, depending upon the pump beam area $S_{ac}=a\cdot 2{\omega }_p$, absorbed pump power $P_{ab}$, thermal load ${\eta }_h$, heat conductivity $k,$ and thermos-optical coefficient $\chi$ of active material [24]. With a high aspect ratio of thermal gain volume $a/{\omega }_p$, the thermally induced aberrations tend to reduce. In the absence of any gain guiding, the thermal lens limits the length of the plane mirrors hybrid resonator to < ${4f}_{th}$ in the fast axis. No thermal gradient exists along the slow axis due to homogeneous top-hat distribution inside the crystal, allowing power scaling without affecting the beam quality. A very high pump intensity of up to 75 kW/cm² can be realized, resulting in a very high gain of crystal [20, 22].

Apart from thermal lensing, thermally induced refractive index distribution along the beam path causes the laser beam to experience different thermal zones, leading to phase shifts that result in net optical path differences (OPD) and subsequent wavefront distortions and limit the beam quality. A study in Ref. [25] highlights that OPD can be significantly reduced by implementing a zigzag trajectory within the InnoSlab structure. The research analyzes the path trajectory of two probe rays (marginal and central) for both pumped InnoSlab and Innozag crystals. It was found that both rays traverse nearly the same thermal zones in Innozag compared to InnoSlab, experiencing similar thermal effects along their profiles, as shown in Fig.4. Additionally, the Innozag approach not only provides higher gain but also results in a 40% reduction in wavefront distortions, thereby enhancing beam quality. This enhancement is especially critical for high-power laser operations, where maintaining higher beam quality is crucial in InnoSlab lasers. Furthermore, wavefront distortions can be minimized in scenarios of complete gain saturation during a single-pass operation.

Fig.4 Simulation of temperature distribution and trajectory of two probe rays from different thermal zones along the propagation length of the slab. (a) InnoSlab, (b) Innozag, and (c) optical path difference along the slab thickness for both InnoSlab (green) and Innozag (orange). The wavefront distortion is six times higher for InnoSlab under similar conditions. (d) Experimental analysis of the effect of wavefront distortion on far-field laser profile confirms the simulation work [25].

Full size|PPT slide

3 Performance analysis of InnoSlab amplifiers

The diode-pumped InnoSlab amplifier represents a power-scalable concept. In most single-pass or multi-pass amplifier configurations, high-output power laser systems face limitations in stored energy extraction due to a constant beam cross-section during its passage through the gain medium, which restricts overall efficiency [34]. Conversely, InnoSlab amplifiers equipped with hybrid resonators enable an increase in the beam cross-section with each successive pass through the gain medium, facilitating efficient power scaling. The amplification factor can be modified by adjusting the gain volume, the number of beam passes, and the cross-section of the amplifying beam within the gain volume. As the width of the crystal increases, the output efficiency and power also tend to increase, allowing for more passes in a single-stage amplification. For a specific crystal width, efficiency enhancement can be achieved by expanding the coverage area of the gain volume. It is feasible to obtain over 90% of the gain volume through precise overlapping of the beam passes within the slab crystal. To determine the total coverage of gain volume in the horizontal direction, one can estimate the beam size using the magnification factor per pass due to beam divergence and by adding up the beam sizes across all passes. Comprehensive details on estimating total gain volume coverage can be found in Ref. [26] for two mirror hybrid resonators and in Ref. [35] for the DP resonator. While ensuring the maximum coverage of pumped crystal during successive passes, a safe distance (security factor) from the side edges of the crystal should be maintained to avoid the diffraction effects. This security factor requires optimization to achieve the best beam characteristics, including output power and beam quality [24]. Similarly, to extract the pump power efficiently and increase the conversion efficiency, the intensity of the seed laser can be enhanced. However, one needs to be careful about the crystal damage threshold. The InnoSlab amplifier typically follows a multi-pass configuration, where the beam traverses multiple times through the gain medium once in single-stage amplification. However, the number of passes is limited due to the risk of the diffraction effects. Thus, a multistage amplification chain can be adopted as a booster to pursue high-power and pulse energy [36, 37]. In a multi-stage ultrafast amplifier, shortening the total beam path inside the slab amplifier (fewer passes) in the first stage and limiting it to one or two passes in the following stages can control the B-integral in the working range. With a proper doping level of the gain medium and given pump intensities, most pump radiation can be absorbed entirely in a single pass and excite the whole gain volume. A dual end-pumping scheme can also be adopted at the amplifier stage to increase the excitation level without increasing the overall pumping intensity [24]. Overall, the InnoSlab amplifier presents a promising solution for high pulse energy lasers operating at high repetition rates. Notably, the design of the cavity does not restrict the repetition rate, unlike traditional time-domain regenerative or multi-pass configurations. This characteristic positions the InnoSlab amplifier as an advantageous concept for ultrafast laser systems in materials processing. Various Nd³⁺, Yb³⁺, and Tm³⁺ doped gain mediums have been demonstrated with InnoSlab amplifiers in different cavity configurations.

3.1 Nd-doped InnoSlab amplifier

Nd-doped materials, particularly Nd:YAG and Nd:YVO₄, are key players in both single and multi-stage amplification, commonly utilized in slab lasers. These materials are favored for their larger emission cross-section and lower saturation energy density, which contribute to efficient energy extraction. The extended upper lifetime of Nd:YAG, at approximately 230 μs, makes it advantageous for energy storage [38], although it tends to support longer pulse durations. In contrast, Nd:YVO₄ excels in generating short pulses at high repetition rates, attributed to its large stimulated emission cross-section, broad pumping wavelength bandwidth, and shorter upper-state lifetime [39]. Nd:YAG also has a higher damage threshold compared to Nd:YVO₄, a quality attributed to its excellent thermo-mechanical properties, making it suitable for high-power laser generation. Traditionally, Nd-doped materials are pumped at 808 nm due to their significant absorption cross-section, but inband pumping has been introduced to reduce quantum defects. Specifically, inband pumping for Nd:YVO₄ is done at 880 nm, while for Nd:YAG at 885 nm/888 nm. The inband pumping method results in a ~30% decrease in quantum defects and mitigates thermal lensing, which is crucial for high-power laser applications. Additionally, Nd:YVO₄, with its relatively large absorption coefficient at 880 nm, offers high absorption efficiency, enabling the use of smaller crystal sizes [40]. Generally, Nd-doped InnoSlab amplifiers operate effectively at room temperature at a picosecond pulse regime. A comprehensive review published in 2015 [23] provided detailed insights into the performance of InnoSlab amplifiers utilizing Nd-doped and Yb-doped materials and covered earlier advancements.

In 2012, a 105 W, 8.4 ps, Nd:YVO₄ InnoSlab amplifier at 120 MHz repetition rate was realized by adopting in-band pumping at 888 nm in a plane−plane cavity. The seed was a 2.8 W passively mode-locked Nd:YVO₄ oscillator emitting 7 ps pulses at 120 MHz. This represents the highest power in the ps regime with Nd:YVO₄ [41]. However, pulse broadening in ps-amplifiers was observed due to the gain narrowing effect. In 2016, Mao et al. [42] amplified 0.4 mJ Q-switched seed laser to pulse energy of 8.4 mJ at 3.6 ns pulse duration and 10 kHz of repetition rate using plane−plane mirror cavity configuration. They achieved a power conversion efficiency of 28% by covering 95% of the pumped area. Later, a multi-mirror resonator for discrete beam path configuration was introduced in InnoSlab amplifiers. In the first description of discrete beam path configuration, a 99 W, 12.4 ps Nd:GdVO₄ amplifier at 81 MHz was realized with a remarkable 42% conversion efficiency. The pulse duration slightly broadened to 12.3 ps from 11.3 ps owing to the finite gain bandwidth of the amplifier [28]. In 2020, this configuration was extended where instead of using plane discrete mirrors, convex mirrors of different radii are used; a 212 W Nd:YAG InnoSlab amplifier was realized with seven single passes by amplifying a 40 W, 11.6 ns Q-switch seed laser. All the other beam parameters remain the same during the amplification [31]. In 2016, a MOPA configuration, based on a ps-SESAM mode-locked oscillator, a regenerative amplifier (RGA) followed by a three-stage Nd:YAG InnoSlab amplifier, was adopted for power and energy scaling. The system amplified a 2 W pre-amplified seed beam at 5 kHz repetition to a pulse energy of 8.2 mJ and an output power of 41 W with 8.5% optical conversion efficiency and >20 amplification factor [37]. Despite being power scalable, these multi-stage designs are complex in dealing with a large number of optical elements involved. A multi-folded design was also applied to simplify the system, where the seed beam is reflected back after one complete pass through the slab. Thus, the beam can be folded two times in a single slab crystal [43]. Here, the seed beam is injected inside the two-plane−plane mirror amplifier cavity, with mirror angles adjusted to change the direction of the amplified beam during successive passes. The beam passes multiple times through the slab, with the incident angle decreasing during each pass and reaching 0° at the final pass. The beam is then reflected back to the same track for double-pass amplification, achieving a 41.6 W output power at an 86 ps pulse duration with 30% extraction efficiency. This is an exciting concept to achieve high power and improve efficiency in a much simpler way. It can be extended to a discrete mirror configuration, giving more liberty to independently adjust the beam path and incident angle for each pass. However, considering double passes in ultrafast pulse amplification, one must be careful about nonlinearities and diffraction effects that rise with high power interaction length. Tab.1 highlights the key performance matrices of the InnoSlab amplifier based on Nd-doped gain mediums over the last decade.

Tab.1 InnoSlab amplifiers based on Nd-doped mediums.

Year	Gain medium	InnoSlab amplifier				o−oefficiency(%)	Number of passes	Remarks
		Power(W)	Energy(mJ)	Repetition rate	Pulse duration
2015 [44]	Nd:YAG	24	4.8	5 kHz	25 ps	5.5/7.5/5.3*	4+3+1	Three-stage
2016 [28]	Nd:GdVO4	99	−	81 MHz	12.4 ps	42	5	DP
2016 [42]	Nd:YVO4	84	8.4	10 kHz	3.4 ns	29.8	8	Single-stage
2016 [37]	Nd:YAG	41	8.2	5 kHz	25.3 ps	4.5/11.9/11.1*	3+2+1	Three-stage
2017 [45]	Nd:YAG	125	25	5 kHz	3 ns	14.3	12	DP
2018 [35]	Nd:YVO4	95	3.1	30 kHz	4.3 ns	28.8	8	DP
2019 [29]	Nd:YAG	212	4.2	50 kHz	7 ns	14.6	12	DP
2020 [31]	Nd:YAG	210	4.2	50 kHz	11.6 ns	19.5	7	DP
2022 [43]	Nd:YAG	41.6	4.1	10 kHz	86.5 ps	29.8	20	Double fold

Note: *Optical conversion efficiency in successive stages.

3.2 kW-level Yb-doped InnoSlab amplifier

High-power ultrafast femtosecond lasers significantly impact both scientific and industrial applications, and the demand for reliable ultrafast sources delivering multi-hundred watts of average power at multi mJ pulse is increasing. Since 2007, the development of Yb-InnoSlab amplifiers has progressed rapidly, driven by experience with Nd-doped amplifiers and the commercial availability of high-brightness LDs in a pumping range of 940−980 nm. In a single-stage Yb:YAG InnoSlab amplifier, nearly transform-limited 636 fs pulses at 620 W average output power and a repetition rate of 20 MHz has been achieved, and further by cascading two amplifiers-stages, this system reached an output power of 1.1 kW and a peak power of 80 MW, with a 615 fs pulse duration. Fig.5 shows the configuration of the two-stage amplifier. The beam takes 7+1 passes through first + second stage amplification, respectively [14]. Here, during the amplification of fs pulses, no spectral broadening by SPM occurs in this geometry, and pulse broadening only occurs due to the gain narrowing effect, thus eliminating the need for external pulse compression. This was the highest output power with a single-stage InnoSlab amplifier in the fs-domain without chirped pulse amplification (CPA). However, the maximum pulse energy with fs-regime remains limited to ~50 μJ.

Fig.5 Dual end-pumped and two-stage configuration for kW-InnoSlab amplifier [14].

Full size|PPT slide

In pursuit of high energy to mJ-level, a CPA-based high-energy Yb:YAG InnoSlab amplifier was demonstrated both in single-stage and multi-stage amplification. In a two-stage CPA-based Yb:YAG InnoSlab amplifier, 1 nJ seed pulses were stretched to ~250 ps by fiber Bragg grating and pre-amplified to 100 μJ before being sent to the main amplifier. With carefully arranged two-stage amplification, highly stable 54 mJ pulses at a 10 kHz repetition rate were achieved [46]. The amplified pulses were recompressed to 1.5 ps at > 500 W output power. Here, about 100 times higher pulse energy was realized from Ref. [14] by adopting the following changes: (i) the repetition of 20 MHz seed oscillators at 1 nJ was tuned from 20 MHz−10 kHz by pulse picker; (ii) in the first stage amplifier, the number of passes was increased from 7 to 9; (iii) in second stage amplifier, the crystal was replaced with an enlarged width of 25 mm and turned to 2-pass. The footprint of the entire amplifier system, including the grating compressor, was approximately 2.4 m × 1 m, categorizing it as one of the most compact high-energy ultrafast systems. The author anticipated tuning the repetition rate to a higher range could enable an output power of up to 1.5 kW with this booster amplifier [47] Recently, Gao et al. [48] adopted an improvised plane−convex hybrid cavity in a CPA-based Yb:YAG InnoSlab amplifier and achieved 33% better results than the plane parallel hybrid cavity. The 417 W CPA system was reported with compressible clean pulses up to 406 fs at 175 kHz. These are the shortest pulses with CPA InnoSlab amplifier [27]. In initial pulse amplification of up to 25 W, a gain narrowing effect occurs and reduces the spectral bandwidth from Δλ = 8.3 to 2.8 nm, which remains constant until 1300 W pump power. A slight spectral broadening occurs at pump power >1300 W, indicating that the InnoSlab amplifier is suitable for producing transform-limited pulses at very high pump power. Furthermore, nonlinear pulse compression techniques have been employed to achieve shorter pulse durations. Pulses up to only a few cycles have also been realized using nonlinear pulse compression in noble gas-filled Herriott multi-pass cells (MPCs). For more details, refer to Section 4.2.

In summary, in InnoSlab geometry for femtosecond pulse amplification at repetition rates above 10 MHz with pulse energies in multi microjoules, CPA is not required, and transform-limited pulses can be obtained without pulse compression. Apart from the gain narrowing effect, no spectral broadening occurs in the pulse spectrum due to the short beam passage through the slab crystal. Tab.2 highlights the key performances of InnoSlab-based amplifiers regarding high average power and pulse duration. It can be observed at high average power and high efficiency; the repetition rate is limited to ≥ 10 kHz. This limit is set by saturation intensity $I_{sat}\left({\lambda }_l\right)=9.7\mathrm{k}\mathrm{W}/{\mathrm{c}\mathrm{m}}^2$ of Yb:YAG [13] and crystal damage threshold. The laser intensity should be well above the saturation intensity for efficient operation. Thus, a repetition rate of multi-kHz is usually required to maintain the energy fluences < 10 J/cm² and avoid optical damage, providing the laser intensity remains constant inside the slab.

Tab.2 InnoSlab amplifiers based on Yb-doped mediums.

Year	Gain medium	InnoSlab Amplifier				o−oefficiency(%)	Number of passes	Remarks
		Power(W)	Energy	Repetition rate	Pulse duration
2009 [24]	Yb:YAG	400	5.3 μJ	75.8 MHz	682 fs	47	9	Single stage
2010 [14]	Yb:YAG	620	31 μJ	20 MHz	636 fs	49.6	7	Single-stage
2010 [14]	Yb:YAG	1100	55 μJ	20 MHz	615 fs	49.6/40.4*	7+1	Two-stage
2011 [49]	Yb:YAG	186	−	26.6 MHz	488 fs	59	9	Single stage
2013 [50]	Yb:YAG	225	450 μJ	500 kHz	5 ps	61	9	Single-stage
2014 [51]	Yb:YAG	320	3.2 mJ	100 kHz	936 fs	−	−	Two-stageCPA
2017 [46]	Yb:YAG	540	54 mJ	10 kHz	1.5 ps	28/23*	9+2	Two-stage CPA
2019 [47]	Yb:YAG	1485	−	46.4 MHz	650 fs	−	7+2	Two-stageCPA
2023 [27]	Yb:YAG	417	2.3 mJ	175 KHz	406 fs	38	9	Single-stage CPA
2024 [48]	Yb:YAG	154	0.88 mJ	175 KHz	519	25.8	5	Single-stage CPA

Note: *Optical conversion efficiency in successive stages.

4 Generation of high peak power few-cycle pulses

Intense few-cycle ultrashort femtosecond laser pulses play crucial roles in many fields, such as time and frequency-domain spectroscopy, attosecond pulse generation via high-harmonics generation, and strong-field physics [52−54]. However, achieving few-cycle fs-pulses simultaneously at high peak and average power is challenging. The Yb-based laser systems offer high average output power in ultrafast pulse regime, and Yb-doped InnoSlab amplifiers have reached the kW level, but pulse duration has remained limited to ~500 fs due to the limited gain bandwidth of Yb:YAG. To overcome these limitations and achieve shorter pulses <100 fs, Yb-systems are usually employed as powerful pump sources for optical parametric chirped pulse amplifiers (OPCPA) or are spectrally broadened via optical non-linear media (post-pulse compression).

4.1 OPCPA

The OPCPA scheme is based on the direct amplification of ultra-broadband pulses in an optical parametric amplifier setup. This method provides advantages such as high gain, a large gain bandwidth, and minimal thermal effects, which contribute to the reduction of pulse duration. Through the process of optical parametric interaction [55], an optical parametric amplification OPA can effectively amplify weak laser signals [56, 57]. On the other hand, the CPA technique boosts the energy of ultrashort pulses by amplifying the temporally stretched pulses in an amplifier, followed by compression, while maintaining the original optical content [58]. Combining the concept of OPA with CPA, known as OPCPA, has opened new pathways for generating ultra-broadband solid-state laser sources. To achieve high efficiency in OPCPA, a stable laser that delivers few-picosecond (~10 ps) pulses with high peak power and good beam quality is essential [59]. The development of Yb-doped kW-level amplifiers at sub-ps to few-ps pulse durations has enabled the building of OPCPA systems delivering energetic ultrashort pulses at desired repetition rates. Particularly three laser geometries: InnoSlab amplifier [60−63], TD-amplifier [127–129], and fiber amplifiers [64, 65] with favorable thermal load management systems have been applied to OPCPA. Tab.3 gives details of the InnoSlab amplifiers that were applied as a pumping source for OPCPA. In 2014, Riedel et al. [62] demonstrated a white light generation (WLG) based OPCPA pumped by adopting a CPA-free 140 W, picosecond Yb:YAG InnoSlab amplifier. The pump and seed source are from the same amplifier. A fraction of pump power ~8 W applied to WL pulse generation in YAG plate via supercontinuum, which will be used as a seed source rather than broadband Ti:Sa laser. The remaining ~130 W was focused in 8 mm BBO for frequency doubling with 60 W at 515 nm as OPCPA pumping. An 11.4 W wavelength-tunable OPCPA delivering 29 fs pulse at 3.25 MHz was realized. Later, in 2019, a NIR wavelength-tuneable high average power OPCPA in KTA was designed for its proposed application in next-generation X-ray light sources, e.g., X-ray free electron lasers (XFELs) [66]. The Yb:YAG pumped Innsolab OPCPA generates a few cycles pulse at a very high peak power of 21 GW. The high repetition, high peak power 880 nm OPCPA generates soft X-rat radiation at >190 eV of photon energy. For details about experimental configuration, see Ref. [67]. The direct amplification of ultra-broadband laser pulses allows this system to achieve outstanding performance. However, the requirement of an intense laser pumping source is crucial to overcome the low efficiency, which adds more complexity and increases the overall cost of the system.

Tab.3 InnoSlab amplifiers as a pumping source for OPCPA.

Year	*Pump source InnoSlab amplifier						OPCA		Remarks
	Gain medium	Power(W)	Repetition rate	Pulse duration		Energy	power	Pulse duration (fs)
2013 [61]	Yb:YAG	500	27.5 kHz	1.39ps		1.39 mJ	−	6.4^	Three-stage OPCPAbrust mode
2014 [62]	Yb:YAG	140	3.25 MHz	1.1 ps		3.5 μJ	11.4	29	Single stage OPCPAwavelength-tunable 700−900
2015 [63]	Yb:YAG	200	500 kHz	1.25 ps		30 μJ	15	20	Single stage OPCPAwavelength-tunable 680−900
2019 [47]	Yb:YAG	661	100 kHz	710 fs		−	88.6	17	Three-stage OPCPA center @ 800 nm
2019 [66]	Yb:YAG	661	100 kHz	626 fs		−	106.2	56.6	wavelength-tunable 1.5−2.0 μm
2020 [67]	Yb:YAG	255	100 kHz	2 ps		−	22.5	9.3	@ 800 nm

Note: *@ Fundamental wavelength 1030 nm, intra-burst repetition rate; ^Supporting spectral bandwidth.

4.2 Non-linear post-pulse compression

The motivation to achieve high peak powers ~GW at ultrashort pulses approaching a few cycles is driven by the abovementioned applications. However, in the direct amplification of ultrafast pulses, several nonlinear effects limit the extent to which peak power can be scaled. Further, the gain narrowing effect increases the pulse duration during the amplification stages. As illustrated in Ref. [14], the amplification of 289 fs seed pulses to kW level, the gain-narrowing effect broadened the pulse duration to 636 fs. Nonlinear post-pulse compression is a promising approach to reaching multi-GW peak powers at high repetition rates without additional power amplification. This technique can shorten the pulse duration to limits beyond the gain bandwidth of the laser active medium via nonlinear external spectral broadening followed by compression while maintaining the high average power [68−71]. In post-compression, pulses are spectrally broadened by inducing self-phase modulation (SPM) in a non-linear medium, which changes the phase of an optical pulse. The total phase change brought in the transmitted pulse from a medium of length $l$ can be estimated as ${\phi }_{\mathrm{N}\mathrm{L}}\left(t\right)=-n_2{\omega }_{\mathrm{o}}I\left(t\right)l/c$ [72]. SPM mainly relies on pulse intensity and propagation length inside the non-linear medium. The shorter pulses are obtained by employing temporal compression in a negative dispersion regime. Different nonlinear pulse compression techniques, such as free propagation in bulk materials [73], multiple-plate continuum generation [74], a planer waveguide, hollow-core fibers (HCF) [75−78], and multi-pass cell (MPC) [69], have been implemented. It is not possible to cover all these techniques; for comprehensive details, refer to Ref. [79]. To recognize the post-pulse compression in the InnoSlab amplifier, 680 fs pulses at 20.8 MHz from 40 W Yb-YAG InnoSlab amplifier are spectrally broadened in 7.1 cm long photonic crystal fiber LMA-35 to output spectrum of Δλ~80 nm. The pulses were compressed to 35 fs in 13 passes between chirped mirrors with −450 fs² GDD per pas [80]. The 600 W output power was attenuated to 40 W to keep the peak power > the damage threshold of LMA-PCF. For high-energy pulses, non-linear mediums with a high damage threshold are required. The MPC-based post-pulse compression, with favorable characteristics such as high compression factor in a more compact structure, high transmission efficiency, and applicability to a wide range of pulse energies, has attracted much attention in recent times.

4.2.1 Multi-pass cell (MPC)

The multi-pass cell (MPC) is a convenient way to increase the optical path length (OPL) by reflecting the beam back and forth inside two or more mirrors, avoiding spatial constraints in optical configuration. This feature makes MPC widely used in gas-sensing technology by laser absorption spectroscopy [81, 82] in a more compact structure. Among various configurations, White-type [83] and Harriot-type [84] MPCs are the most commonly used. While the White-type configuration is more complex, involving multiple mirrors and challenging adjustments, the Harriot-type configuration provides more stability and compactness with only two reflective mirrors and is largely adopted in laser setups. A key feature of Herriot-type MPC is that no beam spot intersection occurs, which could otherwise give rise to interference. The OPL is the main focus of this configuration, which depends upon the type of two mirrors, their focal length, the distance between mirrors, and the incident angle of the input beam [82]. For a comprehensive study of two mirror MPC configurations and trade-offs between different designs, please refer to Ref. [85]. MPC concept can be integrated under multiple circumstances to enhance the laser output performances.

In one scenario, integrating MPC with the existing cavity of mode-locked oscillators increases the pulse energy by extending the resonator length and decreasing the repetition rates from 100 MHz to a few MHz regime at the given average output power ($E_{\mathrm{p}}=P_{\mathrm{o}\mathrm{u}\mathrm{t}}/f_{\mathrm{r}\mathrm{e}\mathrm{p}}$) [86]. As it provides the unity transformation of Gaussian beam q-parameters and keeps the mode size invariant. In this case, the resonator stability remains intact, which otherwise, for any other cavity length extension, can lead to a mismatch between laser modes. The second aspect is the post-pulse compression based on MPC via non-linear spectral broadening in a non-linear medium (bulk or a gas). In the typical nonlinear pulse compression, the laser beam is focused on a single or multiple nonlinear bulk medium. However, one has to be careful of catastrophic damage that can occur due to the self-focusing of the Kerr medium (spatial variation of the Kerr effect) for peak power exceeding the self-focusing threshold of material. This can also cause inhomogeneity in spectral broadening due to large single-pass B-integral, leading to beam quality degradation and limiting transmission efficiency [135]. The self-focusing threshold of glass or other optical crystals is power-dependent quantity rather than pulse intensity and only occurs for beam power $P\gg P_{\mathrm{c}\mathrm{i}\mathrm{t}\mathrm{i}\mathrm{c}\mathrm{a}\mathrm{l}}$. Thus, it can not be managed by scaling the mode size. The critical power of self-focusing is estimated by $P_{\mathrm{c}\mathrm{i}\mathrm{t}\mathrm{i}\mathrm{c}\mathrm{a}\mathrm{l}}=0.146{\lambda }^2/(n_{\mathrm{o}}{n}_2)$ where n_o is material refractive index $n_2$ the non-linear refractive index and ${\lambda }_{\mathrm{l}}$ the lasing wavelength [5]. To achieve spectral broadening without experiencing the other optical pulse shaping effects, the length of the non-linear medium should be small enough that self-focusing occurs outside of the medium. Ignoring the diffraction effects, the self-focusing distance of a beam from the interface of the non-linear optical crystal is given by $z_{\mathrm{s}\mathrm{e}\mathrm{l}\mathrm{f}\text{-}\mathrm{f}\mathrm{o}\mathrm{c}\mathrm{u}\mathrm{s}}=2n_{\mathrm{o}}{\omega }_{\mathrm{o}}^2{\lambda }^2/[\lambda (1/\sqrt{P/P_{\mathrm{c}\mathrm{r}\mathrm{i}\mathrm{t}\mathrm{i}\mathrm{c}\mathrm{a}\mathrm{l}}})]$ [72]. The pulse compression based on MPC with a dielectric medium has proven to be a reliable and efficient technique, where instead of a single-pass focused beam through a single bulk medium, the laser beam repeatedly passes through the thin bulk non-linear medium via MPC. Here, choosing smaller non-linearities per pass < π (~π/10) and increasing total propagation length via MPC, the pulses at peak powers $\gg$ critical self-focusing threshold can be spectrally broadened without beam quality degradation. MPC spectral broadening can be achieved by placing using a solid medium or noble gas as a non-linear medium.

4.2.2 MPC in InnoSlab

In 2016, a technique based on the dielectric MPC-bulk was first proposed [69] in the InnoSlab amplifier for pulse compression. For spectral broadening, a two-stage ~500 W InnSlab amplifier providing 850 fs at a 10 MHz repetition rate and corresponding to 50 μJ pulse energy is coupled to Herriot-type MPC. MPC consists of two concave mirrors of curvature 350 mm. Here, instead of using a separate non-linear medium, the MPC mirror substrate is used as a fused silica medium. The pulse propagates through each 13 mm fused silica 38 times, corresponding to a total propagation length of ~1 m in a non-linear medium. A total power of 446 W was coupled out of MPC with a transmission efficiency of 91% after 18 reflections. In MPC-bulk, the beam quality gets affected at high transmitted power attributed to the induced by thermal and Kerr lines in the mirrors. In Ref. [69], the beam quality remains unchanged up to 400 W of transmitted power; however, M² increases to 1.6 × 1.4 at 446 W from 1.26 × 1.26 at 370 W. The output spectrum of modulated pulses is measured as the width between the points of outer slopes at FWHM. At 375 W, the output spectrum increases to 13.4 nm from 1.6 nm and corresponds to a transform-limited pulse duration of ~170 fs. Later in 2017, the work [87] was extended to achieve a larger compression factor at smaller energy by using the two separate non-linear elements instead of using an MPC mirror substrate and increasing the number of passes to 57. A detailed understanding of the performance mechanism of MPC-bulk towards homogeneous spectral broadening is given in Ref. [87].

The characteristics of MPC in the bulk medium can be further exploited by replacing the nonlinear bulk medium with a noble gas similar to pulse energies in the mJ regime. The delocalization of linear response reduces the variation in SPM across the beam profile, enabling the accumulation of five times higher nonlinear phase per pass and strong spectral broadening than a discrete nonlinear bulk medium. Owing to the small linearities of gases, this scheme is particularly well-suited for energies ranging from 100 μJ to multi-mJ range at high peak powers. This design allows easy adjustment to beam diameter to avoid the ionization effects, and with suitable gas type and pressure, catastrophic self-focusing can be prevented. The gas-filled MPC has demonstrated broad energy ranges from 135 μJ to 200 mJ at sub-50 fs to a few cycles regimes in thin-disk and fiber amplifiers [68, 89−94]. This scheme has also been integrated with an InnoSlab amplifier and supports few-cycle pulses at the mJ level for a larger compression factor of k = 10−40. For the desired compression factor, the nonlinear phase per pass with a given number of passes determines the accumulated nonlinear phase. For example, to compress 600 to 25 fs by a compression factor of about 24, a total non-linear phase of ~12 π is required; hence, a possible high 0.4π nonlinear phase per pass in 30 passes MPC is required for negligible dispersion [88]. Further, if dispersion plays a negligible role, the number of passes and cell length do not solely determine the amount of nonlinear phase. Thus, gas type and pressure allow the adjustment of the spectral broadening characteristics [95]. Similarly, because of the loss-free nature of noble gases and dispersive mirrors, the transmission factor T is only determined by the reflection losses of MPC. Thus, a higher transmission is always realized in MPC-gas than in MPC-bulk. In the first experimental demonstration of gas-filled MPC in InnoSlab [88], pulses with ≥1 mJ energies at 590 fs are spectrally broadened argon-filled Herriot-MPC. The MPCs are adjusted for 58 passes at 3.5 bar Ar-gas pressure and 44 passes at 4 bar. The effect of gas pressure and the number of passes on the compression factor was also observed, as shown in Fig.6. At 3.5 bar gas pressure and 58 passes, the spectrum has broadened to 112 nm, corresponding to a broadening factor of 45. This supports the FTL pulse duration of 25.8 fs, corresponding to a pulse compression factor of 22. Similarly, a compression factor of 20 was achieved by decreasing the number of passes to 44 and increasing the gas pressure to 4 bar.

Fig.6 (a) The experimental set of InnoSlab laser for post-pulse compression in gas-filled MPC with variable gas pressure. (b) The output spectra at different gas pressures and a number of the passes. (c) The autocorrelation of the corresponding spectrally broadened pulse [88]. The table at the bottom represents the summary of parameters obtained in two configurations.

Full size|PPT slide

Further, to compress ps-pulses to a few cycles, a large B-integrtal >200 rad is required in single-stage gas-filled MPC, which can be brought down to <20 rad by cascading the MPC to two stages [95]. The two-stage compression can improve the peak power by setting up the non-equal compression ratio of subsequent stages [96]. A few-cycle pulses of duration 9.6 fs at 6 mJ are demonstrated using two-stage gas-filled MPCs with a record compression factor of 120 [95]. In single-stage gas-filled MPC, the transmission throughput approaches 90%−98%, while in double-stage, it approaches ~70%. This concept is an excellent alternative to hollow-core-based compression and optical parametric amplifiers. It is fully scalable to average power and pulse energy by optimizing factors like gas type, pressure, and foci of imaging mirrors. These factors influence the maximum transmitted pulse energy through MPC.

Tab.4 Nonlinear post-pulse compression of InnoSlab amplifiers.

Year	Gain medium	Input pulse duration	Output power(W)	Outputenergy	Rep. rate	Compressedpulse duration (fs)*	Spectral bandwidth(nm)†	Comp. factor	Remarks
2011 [80]	Yb:YAG	680 fs	23	1 μJ	20.8 MHz	35	80	−	7.1 cm PCF
2106 [69]	Yb:YAG	850 fs	375	37.5 μJ	10 MHz	170	13.5	5	13 mm + 13 mm MPC mirror substrate
2017 [87]	Yb:YAG	700 fs	300	7.5 μJ	40 MHz	99	25	7.5	25 mm+13 mm FS in MPC
2019 [88]	Yb:YAG	590 fs	260	1.04 mJ	250 kHz	26	112	22	Gas-filled MPC 58 passes @ 3.5 bar argon
2020 [97]	Yb:YAG	700 fs	700	7 mJ	100 kHz	70	−	10	14-pass MPC @ 2 bar argon
2021 [98]	Yb:YAG	1200 fs	−	8.6 mJ	1 kHz	44	−	30	34-pass MPC @1.16 bar argon
2023 [95]	Yb:YAG	1200 fs	−	6.7 mJ	1 kHz	10	−	120	Two-stage MPC@1.2 bar argon+0.6 bar neon

Note: *Fourier transform-limited; †Width between outer lobes at FWHM.

5 2 μm InnoSlab laser

The high-power 2 μm lasers are favorable in applications such as laser radar, remote sensing, and material processing due to their low atmospheric absorption. In addition, their strong absorption by water and human tissues makes them extremely useful in laser medicine. The laser operation at 2 μm typically relies on solid-state laser medium-based rare earth doped ions Tm³⁺, Er³⁺, and Ho³⁺ ions doped with different host crystals such as YLF, YAG, and YAP. The Tm³⁺-doped laser materials have a strong and broad absorption band at ~800 nm and can be directly pumped by commercially available LDs. In CW operation alongside 1.06 μm [99−101] and 1.3 μm [102−104], the InnoSlab geometry supporting LD pumping at high intensities makes these laser materials well-suited for 2.0 μm ranges at high power and beam quality [105, 106]. Most of the work on 2 μm with a Tm-doped laser was based on a stable resonator where the beam profile is strongly elliptical [107, 108]. In 2017, the first Tm:YAG InnoSlab laser based on a hybrid resonator, delivering 36.4 W average power with a diffraction-limited beam quality, was demonstrated [106]. Similarly, Tm:YAP is another commonly used host crystal for high-power 2 μm lasers. Due to its twice the thermal conductivity of YLF and twice the emission cross-section of YAG, Tm:YAP has also achieved impressive results with the InnoSlab geometry, reaching 315 W at 1.95 μm in a stable resonator in a stable resonator [109]. In 2023, the first Tm:YLF InnoSlab CW amplifier was demonstrated [110]. A 10 W Tm-doped seed laser with diffraction-limited beam quality was amplified to 40.1 W with 6-pass amplification.

On the other hand, Ho³⁺-doped lasers also emit radiation in the 2 μm region. Here, laser diodes (LDs) at 1.9 μm can be used as a pump source; however, the output power of LDs in this region is typically limited to only a few tens of watts [111]. As a result, Ho³⁺-doped lasers are often pumped by other laser sources, such as Tm:YLF or Tm:YAP, since the emission peak of Tm³⁺-doped lasers around 1.9 μm is well matched to the absorption spectrum of Ho³⁺-doped materials. In the InnoSlab geometry, Ho³⁺-doped media can be efficiently pumped by high-power Tm:YLF or Tm:YAP InnoSlab lasers with stable resonators. This is because, with a stable resonator, the emitted rectangular beam profile (a top-hat distribution in the x-direction and a Gaussian distribution in the y-direction) is well suited to the InnoSlab geometry. Output powers exceeding 100 W have been achieved with Ho:YLF InnoSlab lasers at 2 μm using both stable resonators [112] and hybrid resonators [113] pumped by Tm:YAP slab lasers.

6 Applications

6.1 Industrial applications

The growing industrial market of ultrafast lasers for high-precision material processing has driven progress in developing robust and powerful ultrafast laser systems. The high-power fs-laser at a high repetition rate improves the SNR ratio of measurements [14] and enables maximal precision, processing speed, and efficiency. In material processing with ultrafast laser pulses, the spectral bandwidth-limited pulses are favorable out-of-trouble runtime effects like dispersion of beam transfer and focusing optics. This is the main reason for employing picosecond lasers over femtosecond lasers in industry. However, with shorter pulses having a duration in the range of electro−phonon coupling time, micromachining processes are more stable, and heat-affected zones (HAZ) are minimized. Ultrashort lasers with >100 W average power and a high repetition rate of 100 kHz−100 MHz are required in the high-speed structuring of materials such as metal and ceramics [114]. The InnoSlab amplifiers, being compact and cost-effective and delivering fs-pulses at a high repetition rate with high average power, are favorable in many industrial applications. Here, the pulse energy and repetition rate are governed by the seed laser and can be tailored for the required application. With the flexibility offered in output beam profile (circular, line-shaped: one-dimensional top-hat, and rectangular cross-section: two-dimensional top-hat) and megawatt-level peak powers, the application in glass cutting, drilling, engraving, and material processing has already been demonstrated [115, 116]. Similarly, the nanostructuring on a metal surface using 700−1000 fs InnoSlab laser pulses with a multi-beam approach was demonstrated [117].

6.2 DUV/XUV to THz generations

The improved ultrafast laser systems as driving sources for further nonlinear frequency conversion have enabled the exploration of many new areas with a wide spectral range spanning from the deep-UV to the XUV (200−13 nm) and the mid-IR to the terahertz (THz) range. In scientific research, most efforts to increase the average power of near-IR ultrafast lasers are driven by the generation of coherent extreme UV (XUV) via high harmonic generation HHGs [118]. Thus, researchers would have the unique possibility to unveil the temporal dynamics of fundamental constituents of matter at the nano-level via structural spectroscopy and high-resolution XUV microscopy. Furthermore, HHG also enables the generation of attosecond pulses [119]. In contrast to large-scale synchrotrons for XUV light sources, the ultrafast intense laser driving (HHG) in a noble gas target is a highly attractive source due to the simple table-top setup. High peak intensities in the 10¹³−10¹⁵ W/cm² range are required for nonlinear processes to realize high flux XUV/VUV pulses. This requires laser sources with proportionally higher repetition rates and a corresponding increase in average power. With the repetition rate of the multi-MHz regime, XUV strongly reduces the measuring time and enhances the resolution [147] in applications like ionization, coherent diffractive imaging, and photoelectron spectroscopy [120, 121]. With the recent advancements in ultrafast high-power IR-based Yb-doped InnoSlab amplifiers, especially based on MPC reaching >20 GW peak power, significant progress in HHG systems is anticipated at high-flux XUV, transitioning from kHz to MHz repetition rates. In the first proof of principle experiment, the generation of HHG at tens of MHz was demonstrated in a simple geometry driven by an ultrafast InnoSlab amplifier. Here, 40 W Yb:YAG amplifier delivering pulses 680 fs was externally compressed to 35 fs and was driven to generate HH in a single pass at 20.8 MHz repetition rate and nano-watt power level [80]. In another study, a 270 W InnoSlab amplifier at 59 fs after nonlinear pulse compression was adopted along with resonator enhancement configuration for HHG in xenon for XUV frequency comb [122]. The Yb:YAG InnoSlab amplifier can be efficiently seeded for OPCPA [90] at desired high repetition rates, which is also proven an effective approach for driving HHGs at MHz [123].

In parallel to the progress of high-flux XUV sources via HHG, other research application areas are also emerging in regions of the electromagnetic spectrum of terahertz regime (THz) where laser materials are not well-established or do not exist for the direct generation of short pulses. These THz radiations are located in a less explored yet interesting region of the electromagnetic spectrum that spans from microwave to infrared frequencies. Their lossless penetration in diffractive materials and low energy of the order of several meV as compared to X-rays make them attractive in medical imaging and diagnostic applications. The sources of few or single-cycle THz pulses enable THz time-domain spectroscopy (THz-TDS), a powerful research tool for chemical and molecular spectroscopy [124] or for the study of superconductors in physics [125]. High-power THz radiation sources are limited to accelerator-based sources in a range of 0.1−10 THz with a power level <10 W [126]. However, these systems are very complex, expensive, and have limited accessibility, making THz-TDS very challenging. The laser-based THz sources with high field strength pulse at >100 kHz repetition rates are favorable in next-generation X-ray light sources. Two major laser-based techniques facilitate the intense THz sources: two-color filamentation and optical rectification in electro-optic crystals. In a later technique, the spatial profile of emitted THz pulses follows the shape of a laser pump beam, while cone-shaped THz radiations are emitted from the former one [127]. For optical rectification in electro-optic crystals, the pump source with longer ps-pulses results in higher conversion efficiency in materials like LiNbO₃; however, THz pulses lack the frequency content required in many condensed matter experiments. Therefore, for high field strength THz pulse, a laser system delivering a high average power of multi-100 W with fs-pulses at >100 kHz was designed based on an InnoSlab IR pumping source [97]. For the pumping source, a 100 kHz, 0.84 kW Yb:YAG InnoSlab amplifier was adopted. The amplifier system deliver 700 fs pulses, which were compressed to 70 fs through nonlinear post-pulse compression based on gas-filled MPC. With compressed 344 W of pump power, 144 mW single-frequency THz pulses with 1.44 μJ energy and 150 kV/cm peak field strength were obtained in LiNbO₃. The further increase in pump power 400 W results in LiNbO₃ heating rather than a performance enhancement. The absorption of LibNO₃ at THz frequencies needs to be decreased by improving the crystal cooling mechanism to enhance the power range towards the higher end. The author in Ref. [128] suggests that cooling the crystal down to cryogenic temperatures improves the optical reflection conversion efficiency. In another avenue, benefiting from efficient nonlinear pulse compression, reaching a few cycle durations at high average power becomes possible, placing the InnoSlab as a potential source of isolated attosecond pulses at MHz repetition rates. This breakthrough has significant benefits for various applications in the attosecond domain.

7 Summary

InnoSlab amplifiers, offering a unique combination of efficient thermal management, power scalability, high beam quality, and high repetition rate, have largely contributed to the advancement of high-power ultrafast laser systems. The basic design of InnoSlab amplifier featured a partially end-pumped slab amplifier and a hybrid resonator. It provides a moderate gain factor of 2–10 per pass and an amplification factor of up to ~1000, which can be further improved by adopting a unique zigzag pattern for both the pump and seed beam. Moreover, cascading two or three amplifiers can scale the output power 2−3 times higher; however, adding more amplification stages complicates the system and degrades beam quality due to thermal aberrations. Adopting the double-pass configuration with tight overlapping of the seed beam, instead of using multistage, can improve the amplification factor and scale the power to a higher level with higher efficiency. InnoSlab laser systems have made significant progress in developing high-power amplifiers using both Nd-doped and Yb-doped materials. For ultrafast regeime, Yb-doped materials, in particular, have demonstrated exceptional performance due to their favorable thermal properties and broad emission bandwidth, enabling the generation of high-repetition-rate, ultrafast pulses at the kW level. In these configurations Yb:YAG is widely used gain medium. For instance, in a compact single-stage setup, the highest average power of 600 W at 636 fs and 20 MHz repetition rate was achieved, which was further scaled to an average-power of 1.1 kW with a pulse duration of 615 fs in two-stage amplification. However, pulse energy scaling to a multi-millijoule regime still requires CPA configuration. Apart from Yb:YAG, the other potential gain media such as Yb-doped sesquioxides, tungstates, and ceramic mediums, with their broader gain spectrum and higher thermal conductivity, can further improve the robustness and efficiency of amplifiers. Similarly, integrating InnoSlab amplifiers with post-pulse compression techniques, such as multi-pass cell (MPC) compressors, has further enhanced their capabilities by enabling the generation of few-cycle pulses. The versatility of InnoSlab amplifiers, delivering high peak power at a high repetition rate, extends their applications in extreme ultraviolet (XUV) and terahertz (THz) generation. This ongoing integration of InnoSlab technology with advanced compression techniques and its application in generating XUV and THz radiation underscores its critical role in future ultrafast laser applications.

References

[1] T. Y. Fan and R. L. Byer, Diode laser-pumped solid-state lasers, IEEE J. Quantum Electron. 24(6), 895 (1988)

[2] V. Fomin,V. Gapontsev,E. Shcherbakov,A. Abramov,A. Ferin, D. Mochalov, 100 kW CW fiber laser for industrial applications, in: Proc. 2014 Int. Conf. Laser Opt., 1 (2014)

[3] E. Papastathopoulos, F. Baumann, O. Bocksrocker, T. Gottwald, A. Killi, B. Metzger, S. S. Schad, N. Speker, T. Ryba, and S. Zaske, High-power high-brightness disk lasers for advanced applications, SPIE LASE 116640M, 17 (2021)

[4] L. Liu, S. H. Zhou, Y. Liu, Z. Wang, G. Wang, and H. Zhao, The 5.2 kW Nd:YAG slab amplifier chain seeded by Nd: YVO₄ InnoSlab laser, Chin. Phys. Lett. 34(6), 064202 (2017)

[5] A. V. Smith, B. T. Do, G. R. Hadley, and R. L. Farrow, Optical damage limits to pulse energy from fibers, IEEE J. Sel. Top. Quantum Electron. 15(1), 153 (2009)

[6] J. Zou and X. Lin, High-power laser systems, Laser Photonics Rev. 16(5), 2100741 (2022)

[7] U. Brauch, C. Röcker, T. Graf, and M. Abdou, High‑power, high‑brightness solid‑state laser architectures and their characteristics, Appl. Phys. B 128(3), 58 (2022)

[8] A. Giesen, H. Hügel, A. Voss, K. Wittig, U. Brauch, and H. Opower, Scalable concept for diode-pumped high-power solid-state lasers, Appl. Phys. B 58(5), 365 (1994)

[9] S. Xu, Y. Gao, X. Liu, Y. Chen, D. Ouyang, J. Zhao, M. Liu, X. Wu, C. Guo, C. Zhou, Q. Lve, and S. Ruan, High-repetition-rate and high-power efficient picosecond thin-disk regenerative amplifier, High Power Laser Sci. Eng. 12, e14 (2023)

[10] S. Xu, X. Liu, Y. Gao, Z. Ou, F. Javed, X. He, H. Lu, J. Chen, Y. Chen, D. Ouyang, J. Zhao, X. Wu, C. Guo, C. Zhou, Q. Lue, and S. Ruan, Thin-disk multi-pass amplifier for Kilowatt−Class ultrafast lasers, High Power Laser Sci. Eng. 12, e56 (2024)

[11] M. Ueffing, R. Lange, T. Pleyer, V. Pervak, T. Metzger, D. Sutter, Z. Major, T. Nubbemeyer, and F. Krausz, Direct regenerative amplification of femtosecond pulses to the multi-millijoule level, Opt. Lett. 41(16), 3840 (2016)

[12] A. Penzkofer, Solid state lasers, Prog. Quantum Electron. 12(4), 291 (1988)

[13] W. F. Krupke, Ytterbium solid-state lasers — the first decade, IEEE J. Sel. Top. Quantum Electron. 6(6), 1287 (2000)

[14] P. Russbueldt, T. Mans, J. Weitenberg, H. D. Hoffmann, and R. Poprawe, Compact diode-pumped 1.1 kW Yb:YAG InnoSlab femtosecond amplifier, Opt. Lett. 35(24), 4169 (2010)

[15] P. Russbueldt,T. Mans,D. Hoffmann,S. Schippel, Ultrafast Laser Technology, Springer, Ch. 6, pp 117–134, 2016

[16] J. Löhring, A. Meissner, D. Hoffmann, A. Fix, G. Ehret, and M. Alpers, Diode-pumped single-frequency-Nd:YGG-MOPA for water-vapor DIAL measurements: Design, setup and performance, Appl. Phys. B 102(4), 917 (2011)

[17] J. Löhring,J. Luttmann,R. Kasemann,M. Schlösser,J. Klein,H. D. Hoffmann,A. Amediek,C. Büdenbender,A. Fix,M. Wirth,M. Quatrevalet,G. Ehret, InnoSlab-based single-frequency MOPA for airborne lidar detection of CO₂ and methane, Proc. SPIE, 8959, 92 (2014)

[18] M. Strotkamp, F. Elsen, J. Löhring, M. Traub, and D. Hoffmann, Two-stage InnoSlab amplifier for energy scaling from 100 to >500 mJ for future lidar applications, Appl. Opt. 56(10), 2886 (2017)

[19] K. Du, Optically pumped intensifying agent, in particular a solid intensifying agent, United States Patent, 6351477 B1, 2002

[20] K. Du, N. Wu, J. Xu, J. Giesekus, P. Loosen, and R. Poprawe, Partially end-pumped Nd:YAG slab laser with a hybrid resonator, Opt. Lett. 23(5), 370 (1998)

[21] K . Du, P. Loosen, and R. Poprawe, Optical amplifier arrangement for a solid-state laser, United States Patent, 6654163, 2003

[22] K. Du, Y. Liao, and P. Loosen, Nd:YAG slab laser end-pumped by laser-diode stacks and its beam shaping, Opt. Commun. 140(1-3), 53 (1997)

[23] P. Russbueldt, D. Hoffmann, M. Hofer, J. Lohring, J. Luttmann, A. Meissner, J. Weitenberg, M. Traub, T. Sartorius, D. Esser, R. Wester, P. Loosen, and R. Poprawe, InnoSlab amplifiers, IEEE J. Sel. Top. Quantum Electron. 21(1), 447 (2015)

[24] P. Russbueldt,T. Mans,G. Rotarius,J. Weitenberg,H. D. Hoffmann,R. Poprawe, 400 W Yb:YAG InnoSlab fs-amplifier, Opt. Express 17(15), 12230 (2009)

[25] M. Najafi, . Innozag amplifier: Optimizing the InnoSlab platform to improve beam quality and gain, Opt. Express 32(20), 35652 (2024)

[26] S. Sang, H. Zhang, Y. Mao, X. Zhang, J. Zou, J. Xin, J. Xing, and Y. Jiang, Compact, high-average-power, nanosecond multi-pass Nd:YVO₄ InnoSlab amplifier, Appl. Phys. B 121(2), 131 (2015)

[27] Y. Gao,J. Guo,Y. Huang,Z. Gao,Z. Gan, Z. Tu,X. Liang,R. Li, 417 W, 2.38 mJ InnoSlab amplifier compressible to a high pulse quality of 406 fs, Opt. Lett. 48(20), 5328 (2023)

[28] J. Guo, H. Lin, J. Li, P. Gao, and X. Liang, High power TEM₀₀ picosecond output based on a Nd:GdVO₄ discrete path InnoSlab amplifier, Opt. Lett. 41(12), 2875 (2016)

[29] Q. Gao, F. Javed, H. Zhang, and 2 12 W, Nd:YAG InnoSlab nanosecond laser amplifier, Proc. SPIE 11170, 455 (2019)

[30] Y. F. Mao,H. L. Zhang,J. H. Yuan,X. L. Hao,J. C. Xing, J. G. Xin,Y. Jiang, A high-power diode-pumped Nd:YVO₄ slab amplifier with a hybrid resonator, Laser Phys. Lett. 13(6), 065005 (2016)

[31] F. Javed,H. Zhang,Q. Gao,X. Li,A. Imran, Y. Jiang, A high average power, compact 100 kHz, 11.6 ns Nd:YAG InnoSlab amplifier, Results Phys. 16, 102926 (2020)

[32] K. Z. Han, J. Ning, B. T. Zhang, Y. R. Wang, H. K. Zhang, H. K. Nie, X. L. Sun, and J. L. He, High power single-frequency InnoSlab amplifier, Appl. Opt. 55(20), 5341 (2016)

[33] N. N. Wang,X. L. Wang,T. Zhang,W. Zhang,X. H. Hu, H. Yuan,F. Li,Y. S. Wang,W. Zhao, 23.9 W, 985 fs chirped pulse amplification system based on Yb:YAG rod amplifier, IEEE Photonics J. 11(4), 1 (2019)

[34] D. Li and K. Du, Picosecond laser with 400 W average power and 1 mJ pulse energy, Proc. SPIE 7912, 79120N (2011)

[35] Y. Mao, L. Wang, and H. Zhang, Compact diode-pumped Nd:YVO₄ slab ns-amplifier, Proc. SPIE 10684, 214–220 (2018)

[36] X. Zhang,J. Ye,X. Luo,X. Chen,L. Zhang, X. Xu,H. Ren,Y. Lu,Y. Ma,Q. Gao, J. Sun,W. Wang, 300 W two-stage Nd:YAG InnoSlab microsecond amplifier, Appl. Opt. 60(4), 971 (2021)

[37] Y. Chen,K. Liu,J. Yang,F. Yang,H. Gao, N. Zong,L. Yuan,Y. Lin,Z. Liu,Q. Peng, Y. Bo,D. Cui,Z. Xu, 8.2 mJ, 324 MW, 5 kHz picosecond MOPA system based on Nd:YAG slab amplifiers, J. Opt. 18(7), 075503 (2016)

[38] S. Guy, C. L. Bonner, D. P. Shepherd, D. C. Hanna, A. C. Tropper, and B. Ferrand, High-inversion densities in Nd:YAG: Upconversion and bleaching, IEEE J. Quantum Electron. 34(5), 900 (1998)

[39] D. Krennrich,R. Knappe,B. Henrich,R. Wallenstein,J. A. L’Huillier, A comprehensive study of Nd:YAG, Nd:YAlO₃, Nd:YVO₄ and Nd:YGdVO₄ lasers operating at wavelengths of 0.9 and 1.3 μm. Part 1: Cw-operation, Appl. Phys. B 92(2), 165 (2008)

[40] J. E. Bernard and A. J. Alcock, High-efficiency diode-pumped Nd:YVO₄ slab laser, Opt. Lett. 18(12), 968 (1993)

[41] H. Lin,J. Li,X. Liang, 105 W, <10 ps, TEM₀₀ laser output based on an in-band pumped Nd:YVO₄ InnoSlab amplifier, Opt. Lett. 37(13), 2634 (2012)

[42] Y. Mao,H. Zhang,X. Hao,J. Yuan,J. Xing, J. Xin,Y. Jiang, 8.4 mJ, 10 kHz, 36 ns, Nd:YVO₄ slab amplifier, Opt. Express 24(10), 11017 (2016)

[43] C. Ma, Z. Liu, K. Liu, Y. Yu, X. J. Wang, Y. Bo, D. F. Cui, and Q. J. Peng, High efficiency, 41.6 W, 10 kHz picosecond output based on a Nd:YAG double-pass multi-folded InnoSlab amplifier, Opt. Laser Tech. 148, 107767 (2022)

[44] Y. Chen, F. Q. Li, K. Liu, H. Y. Xu, F. Yang, N. Zong, Y. D. Guo, S. J. Zhang, J. Y. Zhang, Q. J. Peng, Y. Bo, D. F. Cui, and Z. Y. Xu, High-efficiency 2 mJ 5 kHz picosecond green laser generation by Nd:YAG InnoSlab amplifier, IEEE Photonics Technol. Lett. 27(14), 1531 (2015)

[45] Y. F. Mao, H. L. Zhang, J. H. Cui, J. H. Yuan, X. L. Hao, J. Yi, 2 5 mJ, 5 KHz, and 3 ns, Nd:YAG discrete path slab amplifier using a hybrid resonator, Appl. Opt. 56(10), 2741 (2017)

[46] B. E. Schmidt, A. Hage, T. Mans, F. Legare, and H. J. Worner, Highly stable, 54 mJ Yb-InnoSlab laser platform at 0.5 kW average power, Opt. Express 25(15), 17549 (2017)

[47] K. Mecseki, M. K. R. Windeler, A. Miahnahri, J. S. Robinson, J. M. Fraser, A. R. Fry, and F. Tavella, High average power 88 W OPCPA system for high-repetition-rate experiments at the LCLS X-ray free-electron laser, Opt. Lett. 44(5), 1257 (2019)

[48] Y. Gao, J. Guo, Z. Gao, Y. Huang, Z. Tu, and X. Liang, High beam quality chirped pulse amplification laser based on plane−convex hybrid cavity InnoSlab amplifier, Opt. Laser Technol. 168, 109885 (2024)

[49] T. Mans, J. Dolkemeyer, P. Russbüldt, and C. Schnitzler, Highly flexible ultrafast laser system with 260 W average power, Proc. SPIE 7912, 174 (2011)

[50] T. Mans,C. Hönninger,J. Dolkemeyer,A. Letan,C. Schnitzler,E. Mottay, 200 W fs InnoSlab amplifier with 400 μJ pulse energy for industrial applications, Proc. SPIE 8599, 226 (2013)

[51] T. Mans, R. Graf, J. Dolkemeyer, and C. Schnitzler, Femtosecond InnoSlab amplifier with 300 W average power and pulse energies in the mJ-regime, Proc. SPIE 8959, 187 (2014)

[52] L. Von Der Wense, B. Seiferle, M. Laatiaoui, J. B. Neumayr, H. J. Maier, H. F. Wirth, C. Mokry, J. Runke, K. Eberhardt, C. E. Düllmann, N. G. Trautmann, and P. G. Thirolf, Direct detection of the ²²⁹Th nuclear clock transition, Nature 533(7601), 47 (2016)

[53] C. Benko, T. K. Allison, A. Cingöz, L. Hua, F. Labaye, D. C. Yost, and J. Ye, Extreme ultraviolet radiation with coherence time greater than 1 s, Nat. Photonics 8(7), 530 (2014)

[54] F. Labaye,M. Gaponenko,N. Modsching,P. Brochard,C. Paradis,S. Schilt,V. J. Wittwer,T. Sudmeyer, XUV sources based on intra-oscillator high harmonic generation with thin-disk lasers: Current status and prospects, IEEE J. Sel. Top. Quantum Electron. 25(4), 1 (2019)

[55] A. Dubietis, G. Jonušauskas, and A. Piskarskas, Powerful femtosecond pulse generation by chirped and stretched pulse parametric amplification in BBO crystal, Opt. Commun. 88(4−6), 437 (1992)

[56] G. Cerullo and S. De Silvestri, Ultrafast optical parametric amplifiers, Rev. Sci. Instrum. 74(1), 1 (2003)

[57] D. Brida, C. Manzoni, G. Cirmi, M. Marangoni, S. Bonora, P. Villoresi, S. De Silvestri, and G. Cerullo, Few-optical-cycle pulses tunable from the visible to the mid-infrared by optical parametric amplifiers, J. Opt. 12(1), 013001 (2010)

[58] D. Strickland and G. Mourou, Compression of amplified chirped optical pulses, Opt. Commun. 56(3), 219 (1985)

[59] C. Heese, A. E. Oehler, L. Gallmann, and U. Keller, High-energy picosecond Nd:YVO₄ slab amplifier for OPCPA pumping, Appl. Phys. B 103(1), 5 (2011)

[60] M. Schulz, R. Riedel, A. Willner, T. Mans, C. Schnitzler, P. Russbueldt, J. Dolkemeyer, E. Seise, T. Gottschall, S. Hädrich, S. Duesterer, H. Schlarb, J. Feldhaus, J. Limpert, B. Faatz, A. Tünnermann, J. Rossbach, M. Drescher, and F. Tavella, Yb:YAG InnoSlab amplifier: Efficient high repetition rate subpicosecond pumping system for optical parametric chirped pulse amplification, Opt. Lett. 36(13), 2456 (2011)

[61] R. Riedel, M. Schulz, M. J. Prandolini, A. Hage, H. Höppner, T. Gottschall, J. Limpert, M. Drescher, and F. Tavella, Long-term stabilization of high power optical parametric chirped-pulse amplifier, Opt. Express 21(23), 28987 (2013)

[62] R. Riedel, A. Stephanides, M. J. Prandolini, B. Gronloh, B. Jungbluth, T. Mans, and F. Tavella, Power scaling of supercontinuum seeded megahertz-repetition rate optical parametric chirped pulse amplifiers, Opt. Lett. 39(6), 1422 (2014)

[63] M. Puppin,Y. Deng,O. Prochnow,J. Ahrens,T. Binhammer,U. Morgner,M. Krenz,M. Wolf, R. Ernstorfer, 500 kHz OPCPA delivering tunable sub-20 fs pulses with 15 W average power based on an all-ytterbium laser, Opt. Express 23(2), 1491 (2015)

[64] S. Hädrich, S. Demmler, J. Rothhardt, C. Jocher, J. Limpert, and A. Tünnermann, High-repetition-rate sub-5-fs pulses with 12 GW peak power from fiber-amplifier-pumped optical parametric chirped-pulse amplification, Opt. Lett. 36(3), 313 (2011)

[65] P. Rigaud, A. Van de Walle, M. Hanna, N. Forget, F. Guichard, Y. Zaouter, K. Guesmi, F. Druon, and P. Georges, Supercontinuum-seeded few-cycle mid-infrared OPCPA system, Opt. Express 24(23), 26494 (2016)

[66] M. K. R. Windeler,K. Mecseki,A. Miahnahri,J. S. Robinson,J. M. Fraser,A. R. Fry, F. Tavella, 100 W high-repetition-rate near-infrared optical parametric chirped pulse amplifier, Opt. Lett. 44(17), 4287 (2019)

[67] S. Hrisafov, J. Pupeikis, P. A. Chevreuil, F. Brunner, C. R. Phillips, L. Gallmann, and U. Keller, High-power few-cycle near-infrared OPCPA for soft X-ray generation at 100 kHz, Opt. Express 28(26), 40145 (2020)

[68] M. Kaumanns, V. Pervak, D. Kormin, V. Leshchenko, A. Kessel, M. Ueffing, Y. Chen, and T. Nubbemeyer, Multipass spectral broadening of 18 mJ pulses compressible from 13 ps to 41 fs, Opt. Lett. 43(23), 5877 (2018)

[69] J. Schulte, T. Sartorius, J. Weitenberg, A. Vernaleken, and P. Russbudeldt, Nonlinear pulse compression in a multi-pass cell, Opt. Lett. 41(19), 4511 (2016)

[70] M. Benner,M. Karst,P. Gierschke,H. Stark,M. Abdelaal,J. Limpert, High quality pulse post-compression in a multi-pass cell employing enhanced frequency chirping, ASSL, LAC 2023, Vol. 2023, pp 5–6 (2023)

[71] P. Gierschke, C. Grebing, M. Abdelaal, M. Lenski, J. Buldt, Z. Wang, T. Heuermann, M. Mueller, M. Gebhardt, J. Rothhardt, and J. Limpert, Nonlinear pulse compression to 51 W average power GW-class 35 fs pulses at 2 μm wavelength in a gas-filled multi-pass cell, Opt. Lett. 47(14), 3511 (2022)

[72] R. W. Boyd, Nonlinear Optics, 4th Ed., Vol. 2, Elsevier, 2020

[73] M. Seidel, G. Arisholm, J. Brons, V. Pervak, and O. Pronin, All solid-state spectral broadening: An average and peak power scalable method for compression of ultrashort pulses, Opt. Express 24(9), 9412 (2016)

[74] C. H. Lu, W. H. Wu, S. H. Kuo, J. Y. Guo, M. C. Chen, S. D. Yang, and A. H. Kung, Greater than 50 times compression of 1030 nm Yb:KGW laser pulses to single-cycle duration, Opt. Express 27(11), 15638 (2019)

[75] K. F. Mak, M. Seidel, O. Pronin, M. H. Frosz, A. Abdolvand, V. Pervak, A. Apolonski, F. Krausz, J. C. Travers, and P. S. J. Russell, Compressing μJ-level pulses from 250 fs to sub-10 fs at 38 MHz repetition rate using two gas-filled hollow-core photonic crystal fiber stages, Opt. Lett. 40(7), 1238 (2015)

[76] B. H. Chen, M. Kretschmar, D. Ehberger, A. Blumenstein, P. Simon, P. Baum, and T. Nagy, Compression of picosecond pulses from a thin-disk laser to 30 fs at 4 W average power, Opt. Express 26(4), 3861 (2018)

[77] X. Guo, S. Tokita, K. Yoshii, H. Nishioka, and J. Kawanaka, Generation of 300 nm bandwidth 0.5 mJ pulses near 1 μm in a single stage gas filled hollow core fiber, Opt. Express 25(18), 21171 (2017)

[78] L. Lavenu, M. Natile, F. Guichard, Y. Zaouter, M. Hanna, E. Mottay, and P. Georges, High-energy few-cycle Yb-doped fiber amplifier source based on a single nonlinear compression stage, Opt. Express 25(7), 7530 (2017)

[79] T. Nagy, P. Simon, and L. Veisz, High-energy few-cycle pulses: Post-compression techniques, Adv. Phys. X 6(1), 1845795 (2021)

[80] A. Vernaleken, J. Weitenberg, T. Sartorius, P. Russbueldt, W. Schneider, S. L. Stebbings, M. F. Kling, P. Hommelhoff, H. D. Hoffmann, R. Poprawe, F. Krausz, T. W. Hänsch, and T. Udem, Single-pass high-harmonic generation at 208 MHz repetition rate, Opt. Lett. 36(17), 3428 (2011)

[81] Z. Zhang, H. Zhou, X. Chen, and Y. Bi, Design method of variable optical path length multi-pass cell, Appl. Phys. B 127(2), 12 (2021)

[82] J. Wu, T. Grabe, J. L. Götz, J. Trapp, A. S. de Souza, T. Biermann, A. Wolf, P. P. Ley, K. Duan, R. Lachmayer, and W. Ren, Linear scalability of dense-pattern Herriott-type multipass cell design, Appl. Phys. B 129(6), 87 (2023)

[83] J. U. White, Long optical paths of large aperture, J. Opt. Soc. Am. 32(5), 285 (1942)

[84] D. Herriott, H. Kogelnik, and R. Kompfner, Off-axis paths in spherical mirror interferometers, Appl. Opt. 3(4), 523–526 (1964)

[85] A. M. Kowalevicz, A. Sennaroglu, A. T. Zare, and J. G. Fujimoto, Design principles of q-preserving multipass-cavity femtosecond lasers, J. Opt. Soc. Am. B 23(4), 760 (2006)

[86] A. Sennaroglu and J. Fujimoto, Design criteria for Herriott-type multi-pass cavities for ultrashort pulse lasers, Opt. Express 11(9), 1106 (2003)

[87] J. Weitenberg, A. Vernaleken, J. Schulte, A. Ozawa, T. Sartorius, V. Pervak, H. D. Hoffmann, T. Udem, P. Russbüldt, and T. W. Hänsch, Multi-pass-cell-based nonlinear pulse compression to 115 fs at 7.5 μJ pulse energy and 300 W average power, Opt. Express 25(17), 20502 (2017)

[88] P. Russbueldt, J. Weitenberg, J. Schulte, R. Meyer, C. Meinhardt, H. D. Hoffmann, and R. Poprawe, Scalable 30 fs laser source with 530 W average power, Opt. Lett. 44(21), 5222 (2019)

[89] M. Müller, J. Buldt, H. Stark, C. Grebing, and J. Limpert, Multipass cell for high-power few-cycle compression, Opt. Lett. 46(11), 2678 (2021)

[90] L. Lavenu, M. Natile, F. Guichard, Y. Zaouter, X. Delen, M. Hanna, E. Mottay, and P. Georges, Nonlinear pulse compression based on a gas-filled multipass cell, Opt. Lett. 43(10), 2252 (2018)

[91] A. Omar, T. Vogel, M. Hoffmann, and C. J. Saraceno, Spectral broadening of 2 mJ femtosecond pulses in a compact air-filled convex–concave multi-pass cell, Opt. Lett. 48(6), 1458 (2023)

[92] Y. Pfaff, G. Barbiero, M. Rampp, S. Klingebiel, J. Brons, C. Y. Teisset, H. Wang, R. Jung, J. Jaksic, A. H. Woldegeorgis, M. Trunk, A. R. Maier, C. J. Saraceno, and T. Metzger, Nonlinear pulse compression of a 200 mJ and 1 kW ultrafast thin-disk amplifier, Opt. Express 31(14), 22740 (2023)

[93] M. Kaumanns, D. Kormin, T. Nubbemeyer, V. Pervak, and S. Karsch, Spectral broadening of 112 mJ, 1.3 ps pulses at 5 kHz in a LG 10 multipass cell with compressibility to 37 fs, Opt. Letter 46(5), 929 (2021)

[94] C. Grebing, M. Müller, J. Buldt, H. Stark, and J. Limpert, Kilowatt-average-power compression of millijoule pulses in a gas-filled multi-pass cell, Opt. Lett. 45(22), 6250–6253 (2021)

[95] S. Rajhans, E. Escoto, N. Khodakovskiy, P. K. Velpula, B. Farace, U. Grosse-Wortmann, R. J. Shalloo, C. L. Arnold, K. Põder, J. Osterhoff, W. P. Leemans, I. Hartl, and C. M. Heyl, Post-compression of multi-millijoule picosecond pulses to few-cycles approaching the terawatt regime, Opt. Lett. 48(18), 4753 (2023)

[96] E. Escoto,A. L. Viotti,S. Alisauskas,H. Tünnermann,M. Seidel,K. Dudde,B. Manschwetus,I. Hartl,C. M. Heyl, Role of dispersion and compression ratio on the temporal contrast of SPM-broadened post-compressed pulses, CLEO/Europe-EQEC IEEE, pp 1-1 (2021)

[97] P. L. Kramer, M. K. R. Windeler, K. Mecseki, E. G. Champenois, M. C. Hoffmann, and F. Tavella, Enabling high repetition rate nonlinear THz science with a Kilowatt−Class sub-100 fs laser source, Opt. Express 28(11), 16951 (2020)

[98] S. Rajhans,P. K. Velpula,E. Escoto,R. Shalloo,B. Farace,K. Põder,J. Osterhoff,W. P. Leemans,I. Hartl,C. M. Heyl, Post-compression of 8.6 mJ ps-pulses from an Yb:YAG InnoSlab amplifier using a compact multi-pass cell, ASSL AW2A-6 (2021)

[99] T. Xin,J. Cui,Y. Mao,H. Zhang, 205 W laser-diode end-pumped Yb: YAG InnoSlab laser, Laser Phys. 27(6), 065001 (2017)

[100] Y. F. Mao, H. L. Zhang, L. Xu, B. Deng, J. C. Xing, J. G. Xin, and Y. Jiang, An 880-nm laser-diode end-pumped Nd:YVO₄ slab laser with a hybrid resonator, Chin. Phys. Lett. 31(7), 074206 (2014)

[101] L. Cui, H. L. Zhang, L. Xu, J. Li, Y. Yan, C. Duan, P. F. Sha, and J. G. Xin, Laser-diode end-pumped Nd: YVO₄ slab laser under direct pumping into the emitting level, Chin. Phys. Lett. 27(11), 114204 (2010)

[102] X. Li, F. Javed, H. Zhang, X. Liu, T. Chen, S. Yang, T. Zang, Y. Jiang, and J. Jiang, High power diode end-pumped 1.3 μm Nd:YAG InnoSlab laser, Results Phys. 37(2022), 105468 (2022)

[103] S. Zhang, T. Chen, X. Liu, H. Zhang, J. Wang, and H. Guo, Hybrid resonator 1319 nm Nd:YAG InnoSlab laser, Photonics 10(6), 625 (2023)

[104] X. Meng,X. Zhang,X. Chen,X. Luo,J. Ye, L. Zhang,Q. Gao,B. Lu, 170 W Nd:YAG InnoSlab laser at 1319 nm, Opt. Express 31(12), 19126 (2023)

[105] Y. F. Mao and L. Wang, Study of high power Tm: YLF InnoSlab wavelength-selected laser, Laser Phys. 29(11), 115004 (2019)

[106] H. Huang, P. Liu, X. Liu, H. Wang, L. Jin, and D. Shen, Near-diffraction-limited diode end-pumped 2 μm Tm:YAG InnoSlab laser, Laser Phys. Lett. 14(4), 045805 (2017)

[107] M. Schellhorn, S. Ngcobo, and C. Bollig, High-power diode-pumped Tm:YLF slab laser, Appl. Phys. B 94(2), 195 (2009)

[108] J. Li,S. H. Yang,A. Meissner,M. Hofer,D. Hoffmann, A 200 W INNOSLAB Tm:YLF laser, Laser Phys. Lett. 10(5), 055002 (2013)

[109] Y. F. Mao,Q. L. Long,J. M. Gao,Q. C. Wang,Y. Gao, L. Wang, 315 W, 1.94 μm, Tm:YAP InnoSlab laser, Laser Phys. 33(2), 025002 (2023)

[110] J. Gao,Q. Long,L. Wang,Y. Mao,Z. Xu, 40 W high beam quality Tm:YLF InnoSlab amplifier, Laser Phys. 33(2), 025001 (2023)

[111] X. Liu, H. Huang, D. Shen, X. Fan, W. Yao, H. Zhu, J. Zhang, and D. Tang, High-power LD end-pumped Tm:YAG ceramic slab laser, Appl. Phys. B 118(4), 533 (2015)

[112] Q. Wang, Q. Long, Y. Gao, J. Xue, Z. Xu, Y. Mao, and L. Wang, High-efficiency Ho:YLF slab laser with 125 W continuous-wave output power, Appl. Opt. 60(26), 8046 (2021)

[113] Y. F. Mao, Q. L. Long, and J. M. Gao, Hundred-watts-level Ho:YLF slab laser with good beam quality, Laser Phys. 34(5), 055001 (2024)

[114] N. Hodgson,M. Laha,T. S. Lee,H. Haloui,S. Heming,A. Steinkopff, Industrial ultrafast lasers − systems, processing fundamentals, and applications, CLEO 2019, pp 1–2, 2019

[115] K. Du, Unique performances and favourable applications of InnoSlab lasers, Proc. SPIE 7193, 607 (2009)

[116] D. Wortmann, T. Mans, and J. Weitenberg, Multi-100 W average power fs-laser for material processing applications, ICALEO 2009 2009(1), 856 (2009)

[117] C. Schnitzler, T. G. Mans, J. Dolkemeyer, and P. Dittmann, High power, high energy, and high flexibility: Powerful ultrafast lasers based on InnoSlab technology, Proc. SPIE 10911, 1091103 (2019)

[118] J. G. Eden, Erratum: High-order harmonic generation and other intense optical field-matter interactions: Review of recent experimental and theoretical advances, Prog. Quantum Electron. 29(3-5), 257 (2005)

[119] R. Locher, L. Castiglioni, M. Lucchini, M. Greif, L. Gallmann, J. Osterwalder, M. Hengsberger, and U. Keller, Energy-dependent photoemission delays from noble metal surfaces by attosecond interferometry, Optica 2(5), 405 (2015)

[120] C. H. Zhang and U. Thumm, Attosecond photoelectron spectroscopy of metal surfaces, Phys. Rev. Lett. 102(12), 123601 (2009)

[121] A. Cingöz, D. C. Yost, T. K. Allison, A. Ruehl, M. E. Fermann, I. Hartl, and J. Ye, Direct frequency comb spectroscopy in the extreme ultraviolet, Nature 482(7383), 68 (2012)

[122] J. Moreno, F. Schmid, J. Weitenberg, S. G. Karshenboim, T. W. Hänsch, T. Udem, and A. Ozawa, Atomic physics toward XUV frequency comb spectroscopy of the 1 S−2 S transition in He⁺, Eur. Phys. J. D 77(4), 67 (2023)

[123] S. Hädrich, J. Rothhardt, M. Krebs, F. Tavella, A. Willner, J. Limpert, and A. Tünnermann, High harmonic generation by novel fiber amplifier based sources, Opt. Express 18(19), 20242 (2010)

[124] K. Ajito and Y. Ueno, THz chemical imaging for biological applications, IEEE Trans. Terahertz Sci. Tech. 1(1), 293 (2011)

[125] M. C. Nuss, K. W. Goossen, J. P. Gordon, P. M. Mankiewich, M. L. O’Malley, and M. Bhushan, Terahertz time-domain measurement of the conductivity and superconducting band gap in niobium, J. Appl. Phys. 70(4), 2238 (1991)

[126] B. Green, S. Kovalev, V. Asgekar, G. Geloni, U. Lehnert, . High-field high-repetition-rate sources for the coherent THz control of matter, Sci. Rep. 6(1), 22256 (2016)

[127] V. Y. Fedorov and S. Tzortzakis, Powerful terahertz waves from long-wavelength infrared laser filaments, Light Sci. Appl. 9(1), 186 (2020)

[128] S. W. Huang, E. Granados, W. R. Huang, K. H. Hong, L. E. Zapata, and F. X. Kärtner, High conversion efficiency, high energy terahertz pulses by optical rectification in cryogenically cooled lithium niobate, Opt. Lett. 38(5), 796 (2013)

Declarations

The authors declare that they have no competing interests and there are no conflicts.

Author contributions

All authors have accepted responsibility for the entire content of this manuscript and approved its submission.

Acknowledgements

We are thankful to the National Key Research and Natural Science Foundation of China (Nos. 62105225, 62275174, 61975136, and 61935014), the Development Program of China (No. 2022YFB3605800), the Natural Science Foundation of Top Talent of Shenzhen Technology University (No. GDRC202106), Shenzhen University Stability Support Project (Nos. 20220719104008001 and 20220718173849001), and Guangdong Provincial Engineering Technology Research Center for Materials for Advanced MEMS Sensor Chip (No. 2022GCZX005).

RIGHTS & PERMISSIONS

2025 Higher Education Press