|
|
1.INTRODUCTIONThere are several approaches to increasing the energy and average power of a femtosecond laser oscillator. The most common and straightforward approach is to use amplifiers such as fiber, innoslab and thin-disk amplifiers [1-4]. The use of amplifiers adds additional complexity, noise and, not least, costs. Another approach is to use enhancement cavities which allow “storage” of the laser radiation inside of a passive cavity [5, 6]. The range of applications for enhancement cavities is rather narrow due to the fact that they can tolerate only small losses. So far these cavities are mostly used for high harmonic generation, a low efficiency (losses) nonlinear process. Enhancement cavities require precise locking of the repetition rate of the oscillator and its carrier envelope offset frequency (CEO) which adds additional complexity to the system [7]. A third possibility is to directly scale the power and energy of the laser oscillator. This route demonstrated exciting progress over the recent years with peak powers in the range 10-100 MW, obtainable directly from oscillators, which corresponds to even higher intra-cavity peak powers, typically by a factor 5-10 [8-12]. Interestingly, such high peak powers are achieved in the anomalous dispersion regime (ADR), thus the scaling potential of the normal dispersion regime (NDR) was not even reached (in contrast to fiber oscillators) [13]. These high peak powers are combined with high average power in the range of few hundreds of watts. Moreover, carrier envelope phase (CEP) stabilization of a thin-disk oscillator was performed recently. Efficient external spectral broadening and compression resulted in 10 fs pulse duration and 10 W average power[14]. Such unique parameters of femtosecond thin-disk oscillators open doors to new exciting applications in research and industry. 2.KERR-LENS MODE-LOCKED THIN-DISK OSCILLATORThe first SESAM mode-locked thin-disk oscillator was demonstrated in 2000 [15] and the first KLM thin-disk oscillator in 2011[11]. Despite of this “late” realization the progress in the development of KLM oscillators is incredibly fast. The first KLM thin-disk oscillator was demonstrated at 4 MW peak power and considerably shorter pulses as compared to SESAM mode-locked oscillators. Recent experiments performed in our lab demonstrate power scaling up to 30 MW [16]. Another group from Japan reported on about 50 MW (almost 1GW intra-cavity peak power) [17]. 2.1Cavity designIn order to make the analysis brief we refer to the cavity design of the Ti:Sa oscillator and numerous experimental and theoretical works published in the context of that development [18-22]. Here, we emphasis the main difference in the cavity design between the standard Ti:Sa oscillator and the thin-disk KLM oscillator. These main differences are: a) spatial separation of the Kerr medium from the gain medium, b) cavity design (convex-concave) leading to large spot sizes over all elements in the cavity. These differences are schematically illustrated in Fig. 1. The point a) makes it possible to change the nonlinearity in the Kerr medium by varying the thickness and medium type (in other words the nonlinear refractive index n2). In contrast to the situation in Fig. 1a this procedure does not influence the absorption in the gain medium or oscillator efficiency. Additionally an increase of the mode size in the Kerr medium is possible by exchanging mirrors R1 and R2. Point b) helps to reduce constraints on the damage threshold of the intra-cavity optics and to realize a compact oscillator without the necessity for a longer cavity. Figure 1.Rough graphical representation of the main differences between standard Ti:Sa X shape cavity (left) and the resonator of a KLM thin-disk oscillator (right). R1,R2 concave mirrors; K, Kerr medium; HD, high dispersive mirror; H, hard aperture; OC, output coupler. The cavity configuration is from [16]  2.2Hard and soft aperture optimizationIt is important to understand that hard and soft apertures KLM are generally not independent. The mode size at the position of the hard aperture depends on the mode size at the position of the soft aperture (gain). This coupling may result in much more complex saturation characteristics of the self-amplitude (SAM) coefficient of KLM. It is possible to suppress this coupling to some extent by choosing the resonator configuration in a way that the power dependent mode change in the gain medium is less pronounced than at the position of the hard aperture. Indeed, it is more beneficial to maximize both soft and hard aperture effects to obtain highest modulation depth. Experimental optimization of the hard aperture is simple and implies checking the influence of different aperture sizes on the KLM start-up, pulse duration and optical-to-optical efficiency. The aperture H diameter in Fig.1b is 4 mm and was optimized this way. 2.3Advantages of the KLM technique.A detailed analysis of the experimental results obtained for nearly the same cavity configuration with two different mode-locking techniques is presented in [23] and summarized in Table 1. A practical advantage of the KLM technique, especially in the development phase, is the ability to tune different parameters governing soliton mode-locking by simple manipulation of the cavity length, Brewster plate (Kerr medium) and hard aperture (drilled metal plate). Table 1.Summary on disadvantages and advantages of two different mode-locking techniques.
3.ENERGY SCALING CONCEPT3.1Average power scalingThe main advantage of the thin-disk concept is power scalability [24]. It relies on an increase of the mode area in the gain medium proportionally to an increase of the pump power. This geometry is proven to provide high average powers and recently resulted in the nearly TEM00 operation at 1 kW level [25]. 3.2Experimental resultsThe main idea behind the energy scaling of KLM oscillator is a mode size increase in the Kerr medium. The onset of instabilities (like Q-switching, multiple pulses, CW breakthrough) is caused by the excessive nonlinearity arising when the peak power in the oscillator reaches some threshold value Pmax. An increase of the mode size in the Kerr medium situates the onset of the excessive nonlinearity at higher intra-cavity values of Pmax. This simple approach has been recently demonstrated experimentally in our group. Additionally to the increase of the mode diameter in the Kerr medium up to ≈120 μm, the group delay dispersion (GDD) was increased up to -48000 fs2 and the number of passes through the disk was doubled. These measures resulted in 230 W average output power, 330 fs pulse duration and 11.5 μJ pulse energy which correspond to 30 MW peak power. Figure 2.Cavity configuration (not to scale): OC: output coupler with 20% transmission, TD: Yb:YAG thin-disk, HD: high dispersive mirrors with ~ -3000 fs2 per bounce, FM1,FM2: focusing mirrors with f=450 mm, KM: 1 mm thick sapphire-plate, HR: high reflectivity mirrors. b) Autocorrelation-trace, spectrum and corresponding beam-profile. The picture is taken from [16].  3.3Main limitationsThe theoretical limitations of the maximal achievable pulse energy from mode-locked oscillators are definitely not yet reached. The main limitations are originating from the experimental and technical side. These limitations can be summarized as: a) thermal effects in the intra-cavity optics, especially dispersive mirrors, b) thermal effects in the thin-disk gain medium, c) high sensitivity to the thermal effects at the stability edge, d) rather moderate modulation depth (5%) of the current configuration of the KLM resonator, c) nonlinearity of air for peak powers exceeding 300 MW 3.4Possible industrial useIndustrial applications require incredible long term stability and reproducibility. This, in turn, requires serious engineering efforts to be done for KLM oscillators. Among these engineering efforts crucial ones are: a) thermal management of the laser housing and all optics and opto-mechnical parts, b) robust and mechanically stable housing, c) prevention of air flow inside the oscillator cavity, d) ultra-stable opto-mechanics. 4.FUTURE PROSPECTSBased on the simple geometrical scaling concept of the KLM thin-disk oscillator and its intrinsically stable operation, CEP stabilized oscillators with average power levels of >300 W and peak powers >300 MW come into reach. Implementation of spectral broadening and compression will result in few-cycle pulses with an order of magnitude increased peak power (3 GW). These unique parameters will strongly expand the range of applications for these oscillators and turn many now unfeasible experiments into reality. REFERENCEST. Eidam, S. Hanf, E. Seise et al.,
“Femtosecond fiber CPA system emitting 830 W average output power,”
Opt Lett, 35
(2), 94
–6
(2010). https://doi.org/10.1364/OL.35.000094 Google Scholar
P. Russbueldt, T. Mans, J. Weitenberg et al.,
“Compact diode-pumped 1.1 kW Yb:YAG Innoslab femtosecond amplifier,”
Opt Lett, 35
(24), 4169
–71
(2010). https://doi.org/10.1364/OL.35.004169 Google Scholar
T. Metzger, A. Schwarz, C. Y. Teisset et al.,
“High-repetition-rate picosecond pump laser based on a Yb:YAG disk amplifier for optical parametric amplification,”
Opt Lett, 34
(14), 2123
–5
(2009). https://doi.org/10.1364/OL.34.002123 Google Scholar
J. P. Negel, A. Voss, M. A. Ahmed et al.,
“1.1 kW average output power from a thin-disk multipass amplifier for ultrashort laser pulses,”
Optics Letters, 38
(24), 5442
–5445
(2013). https://doi.org/10.1364/OL.38.005442 Google Scholar
C. Gohle, T. Udem, M. Herrmann et al.,
“A frequency comb in the extreme ultraviolet,”
Nature, 436
(7048), 234
–7
(2005). https://doi.org/10.1038/nature03851 Google Scholar
R. J. Jones, K. D. Moll, M. J. Thorpe et al.,
“Phase-coherent frequency combs in the vacuum ultraviolet via high-harmonic generation inside a femtosecond enhancement cavity,”
Phys Rev Lett, 94
(19), 193201
(2005). https://doi.org/10.1103/PhysRevLett.94.193201 Google Scholar
I. Pupeza, S. Holzberger, T. Eidam et al.,
“Compact high-repetition-rate source of coherent 100 eV radiation,”
Nature Photonics, 7
(8), 608
–612
(2013). https://doi.org/10.1038/nphoton.2013.156 Google Scholar
C. J. Saraceno, F. Emaury, O. H. Heckl et al.,
“275 W average output power from a femtosecond thin disk oscillator operated in a vacuum environment,”
Optics Express, 20
(21), 23535
–23541
(2012). https://doi.org/10.1364/OE.20.023535 Google Scholar
C. J. Saraceno, F. Emaury, C. Schriber et al.,
“Ultrafast thin-disk laser with 80 muJ pulse energy and 242 W of average power,”
Opt Lett, 39
(1), 9
–12
(2014). https://doi.org/10.1364/OL.39.000009 Google Scholar
D. Bauer, I. Zawischa, D. H. Sutter et al.,
“Mode-locked Yb:YAG thin-disk oscillator with 41 microJ pulse energy at 145 W average infrared power and high power frequency conversion,”
Opt Express, 20
(9), 9698
–704
(2012). https://doi.org/10.1364/OE.20.009698 Google Scholar
O. Pronin, J. Brons, C. Grasse et al.,
“High-power 200 fs Kerr-lens mode-locked Yb:YAG thin-disk oscillator,”
Opt Lett, 36
(24), 4746
–8
(2011). https://doi.org/10.1364/OL.36.004746 Google Scholar
A. Amani Eilanlou, Y. Nabekawa, M. Kuwata-Gonokami et al.,
“Kerr lens mode-locking of a high-average-power thin-disk ring oscillator,”
OSA Technical Digest, Google Scholar
V. L. Kalashnikov, Chirped-Pulse Oscillators: Route to the Energy-Scalable Femtosecond Pulses, InTech, (2012). Google Scholar
O. Pronin, M. Seidel, F. Lücking et al., Next-generation source of waveform-controlled light,
(2014). Google Scholar
J. Aus der Au, G. J. Spuhler, T. Sudmeyer et al.,
“16.2-W average power from a diode-pumped femtosecond Yb:YAG thin disk laser,”
Opt Lett, 25
(11), 859
–61
(2000). https://doi.org/10.1364/OL.25.000859 Google Scholar
J. Brons, V. Pervak, E. Fedulova et al., Power-scaling a Kerr-lens mode-locked Yb:YAG thin-disk oscillator via enlarging the cavity mode in the Kerr-medium,
(2014). Google Scholar
N. Kanda, A. A. Eilanlou, T. Imahoko et al.,
“High-Pulse-Energy Yb:YAG Thin Disk Mode-Locked Oscillator for Intra-Cavity High Harmonic Generation,”
OSA Technical Digest (online)., Google Scholar
D. E. Spence, P. N. Kean, and W. Sibbett,
“60-Fsec Pulse Generation from a Self-Mode-Locked Ti-Sapphire Laser,”
Optics Letters, 16
(1), 42
–44
(1991). https://doi.org/10.1364/OL.16.000042 Google Scholar
F. Krausz, M. E. Fermann, T. Brabec et al.,
“Femtosecond Solid-State Lasers,”
Ieee Journal of Quantum Electronics, 28
(10), 2097
–2122
(1992). https://doi.org/10.1109/3.159520 Google Scholar
T. Brabec, and F. Krausz,
“Intense few-cycle laser fields: Frontiers of nonlinear optics,”
Reviews of Modern Physics, 72
(2), 545
–591
(2000). https://doi.org/10.1103/RevModPhys.72.545 Google Scholar
G. Cerullo, S. Desilvestri, and V. Magni,
“Self-Starting Kerr-Lens Mode-Locking of a Ti-Sapphire Laser,”
Optics Letters, 19
(14), 1040
–1042
(1994). https://doi.org/10.1364/OL.19.001040 Google Scholar
M. Piche,
“Beam Reshaping and Self-Mode-Locking in Nonlinear Laser Resonators,”
Optics Communications, 86
(2), 156
–160
(1991). https://doi.org/10.1016/0030-4018(91)90552-O Google Scholar
O. Pronin, Towards a Compact Thin-Disk-Based Femtosecond XUV Source, Springer(2013). Google Scholar
A. Giesen, and J. Speiser,
“Fifteen years of work on thin-disk lasers: Results and scaling laws,”
Ieee Journal of Selected Topics in Quantum Electronics, 13
(3), 598
–609
(2007). https://doi.org/10.1109/JSTQE.2007.897180 Google Scholar
Y. H. Peng, Y. X. Lim, J. Cheng et al.,
“Near fundamental mode 1.1 kW Yb:YAG thin-disk laser,”
Optics Letters, 38
(10), 1709
–1711
(2013). https://doi.org/10.1364/OL.38.001709 Google Scholar
|
