Open Access
Add to BibSonomy
Abstract
We generate bandwidth limited 10 µJ pulses of 92 fs pulse width using an adaptive fiber Bragg grating stretcher (FBG) in conjunction with a Lyot filter. The temperature controlled FBG is used to optimize the group delay, whereas the Lyot filter counteracts gain narrowing in the amplifier chain. Soliton compression in a hollow core fiber (HCF) allows for access to the few-cycle pulse regime. Adaptive control further enables the generation of nontrivial pulse shapes.
© 2023 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
Ultrafast Yb fiber amplifiers have reached a high level of maturity and are now routinely being used for micromachining in different industrial settings as well for high-field physics research, such as nuclear spectroscopy [1], just to name one example. State of the art systems can now deliver pulses with energies of 10 mJ and average powers exceeding 10 kW with a pulse width as short as 106 fs [2,3]. Refs. 1 and 2,3 are good examples of the two main system configurations typically employed for the construction of high power Yb fiber amplifiers. In ref. [1] a nonlinear fiber power amplifier is used to simultaneously spectrally broaden and amplify fs input pulses, based on the use of near parabolic pulse formation [4]; in refs. [2,3], the amplifier system is designed to minimize any spectral broadening in the power amplifier; for simplicity we refer to such amplifier systems as bandwidth preserving. To obtain ≈100 fs pulses or shorter is straight-forward with the first system configuration, whereas in the second system configuration a rather sophisticated pulse shaping system needs to be employed to counter-balance gain narrowing in the amplifier chain [3]. A major restriction of nonlinear fiber amplifier systems is the fact that energy scaling via the implementation of chirped pulse amplification is very limited, as the amplified pulse quality degrades rapidly when the amplified pulse width is longer than just a few ps, whereas bandwidth-preserving fiber amplifier systems allow the use of pulse stretching to a few ns without degradation of pulse quality for a properly designed stretcher/compressor pair.
In technology there still remains the need for a relatively simple and scalable bandwidth preserving Yb fiber amplifier system that exploits the full Yb amplification bandwidth to reach the 100 fs pulse regime. Such short pulses reduce the pulse energy requirements for pulse compression [5–7] and UV generation in hollow fibers [8–10], greatly increasing the practicability of such systems. We show here that we can adaptively control the output pulses of a compact bandwidth-preserving chirped pulse amplification system via a pulse shaper based on thermal control of a fiber Bragg grating (FBG) stretcher [11–13]. With insertion of a simple Lyot filter for amplified spectral width management we generate pulses as short as 87 fs with pulse energies up to 5 µJ and 92 fs for a pulse energy of 10 µJ. We further demonstrate that the FBG stretcher allows for the generation of complex pulse shapes such as pulse trains, as relevant for ablation cooled micromachining [14]. Finally, we compress the laser output to 11.7 fs via using a gas-filled single-ring hollow-core photonic crystal fiber (HC-PCF) for soliton-effect self-compression.
2. Experiment
The experimental set-up is shown in Fig. 1. We use an IMRA ultrafast Yb fiber oscillator [1] based on a nonlinear amplifying loop mirror design [15,16] delivering 80 fs (stretched to a few hundred fs) at a repetition rate of 75 MHz with an output power of 2 mW as the seed laser. The oscillator spectrum is shown in Fig. 2. The Yb oscillator is amplified in three stages of single-mode Yb pre-amplifiers and a final large mode area Yb power amplifier based on a 1.8 m length of Yb doped leakage channel fiber [17] with a core diameter of 35 µm. All fibers were polarization maintaining. We insert an adaptive FBG stretcher [11] with nominal 2nd, 3rd and 4th order dispersion values matched to the dispersion of the whole system after the first pre-amplifier. The FBG stretcher has a bandwidth of 54 nm centered at 1060 nm and a dispersion of +6 ps2 at 1060 nm. An acousto-optic down-counter after the 2nd pre-amplifier reduces the repetition rate to 1 or 2 MHz. A Lyot filter producing a loss of 85% at the center of the amplified spectrum and a bandwidth of 30 nm is inserted after the 3nd pre-amplifier and used to maximize the amplified spectral width in the power amplifier. The Lyot filter consists of a 0.15 mm thick YVO4 $\theta = 45^\circ \; \; \; \varphi = 0^\circ $ crystal placed between two polarizers. The relative angle between the polarizers and the crystal controls the spectrum through the Lyot filter. The angles are chosen through an iterative process to maximize the bandwidth from the power amplifier while checking that there is sufficient seed light for the amplifier. A halfwave plate after the Lyot filter insures the seed light has the correct polarization for the amplifier. The leakage channel power amplifier produced a high mode quality at the output and we verified an M2 value of 1.1 × 1.2 at an average output power of 18 W. A bulk grating pair compressor with groove density of 1600 l/mm and a dispersion of around -6.5 ps2 completes the system.
Fig. 2. Red: Oscillator spectrum. Blue: spectrum into Lyot filter. Yellow: Spectrum out of Lyot filter. Green: amplified pulse spectrum. All spectra are normalized.
The adaptive FBG stretcher has a length of 7 cm and is thermally contacted to 32 resistive heater elements, which allow for group delay adjustments between 1–2 ps for each heated FBG section. The system is similar to Lee et al. [13], with the heaters mounted on a circuit board, and the driver interfaced to a computer for manual control, or automated optimization such as by neural networks. In this case, the stretcher was open to air, but more temperature stabilization and isolation can be used for sensitive systems such as shaped supercontinuum generation [18].
The Lyot filter is used to decrease the spectral intensity in the center of the spectrum relative to the edges, which leads to a broader spectrum after amplification allowing for shorter pulses. The spectrum after the Lyot filter is also shown in Fig. 2. For simplicity we only use a single-stage Lyot filter in the set-up; we have also modelled any bandwidth improvement possible with a 2 or 3 stage Lyot filter, which can in principle increase the amplified bandwidth further, though with the trade off of even larger excesses losses added to the system, and an increase in amplified spontaneous emission.
The temporal profiles, obtained by retrieving the spectrograms measured using a commercial FROG (second harmonic frequency resolved optical gating) device are shown in Fig. 3. By optimizing the heater configuration for pulses with 5 µJ pulse energy and 1 MHz repetition rate, we achieved a full-width-half-maximum (FWHM) duration of 87 fs, which we believe is a record for bandwidth-preserving Yb fiber amplifiers. When increasing the average power to 10 W and the pulse energy to 10 µJ, we obtain pulses with a FWHM of around 144 fs, with greatly reduced pulse quality as shown in Fig. 3(b). After re-optimization of the FBG setting the pulse width is reduced to 92 fs, as seen in Fig. 3(c), clearly demonstrating the ability of adaptive control to ensure the generation of near bandwidth-limited pulses with a change in laser power. The pulse pedestal energy content for Fig. 3(a), 3(b), and 3c are 29%, 44%, and 29% respectively. The FBG heater configuration for the two cases is shown in Fig. 4(a) and Fig. 4(b). Here a value of 0 corresponds to no current applied to the resistive heaters and a value of 1 corresponds to the maximum allowable average current (39 mA) per heater element. The heaters are mainly acting on the blue side of the pulse spectrum, exemplifying the good dispersion compensation even without engaging adaptive control. The change in dispersion between 5 and 10 µJ can be mainly attributed to the difference in self-phase modulation (SPM), which was calculated to amount to around 2π and 4π for the 5 and 10 µJ pulses respectively. The residual satellite pulses observable in Figs. 3(a) and 3(b) are attributed mainly to imperfection in the dispersion profile of the FBG as well as small departures from an ideal thermal profile of the grating heaters, which was not optimized in the present work. Also, any perturbations to the FBG in process of mounting it to the heating elements can degrade the pulse quality and increase the pedestal content, as we observed experimentally. In the future resistive heating element directly attached to the bare fiber should allow improved heating profile control with a higher density of heating elements along with a minimization of fiber perturbations, which will be very beneficial for improving the pulse quality further.
Fig. 3. Red traces: Temporal pulse data shown in red for a: the optimized 5 µJ pulse, b: the unoptimized 10 µJ pulse, c: the optimized 10 µJ pulse. Blue traces: fitted gaussian shaped pulses.
Fig. 4. Heater settings for a: the optimized 5 µJ pulse, b: the optimized 10 µJ pulse, c: the pulse train.
In principle adaptive FBGs should allow compensation of even higher values of SPM, however, in this case we observed an onset of pulse energy shedding into a broad pulse background induced by spectral ripple and the presence of satellite pulses [19] due to small misalignments between concatenated polarization maintaining fibers. In principle a larger number of heater elements with a higher spatial resolution can minimize such effects. On the other hand, scaling to higher average powers with the present system is no problem when increasing the repetition rate accordingly. Hence an average power of 100 W is reachable for a pulse repetition rate of 10 MHz. Since the fabrication of FBGs with a length up to 30 cm is quite feasible, the pulse energies can in principle be scaled by up to a factor of 4 or even 8 by the implementation of one or two of such long stretcher FBGs. Even higher powers can be reached via coherent combination [2].
FBG control can in principle be automated via a look-up table or via a neural network. Moreover, the heaters also allow for the generation of non-trivial pulse shapes, as shown in Fig. 5. Here the heaters and compressor were adjusted to produce a pulse train of 4 pulses in a 3 ps window. The corresponding heater profile for the generation of a 4 pulse burst is shown in Fig. 4(c). An increase in the variation of the heater profile with heater number is noticeable compared to Fig. 4(a), however, clearly the heater profile is not perfectly modulated as expected, due to the different functionalities of the adaptive FBG that are being fulfilled at the same time. Pulse bursts may be interesting for ablation cooled micromachining applications [14], the extension of the burst length is possible in principle by the concatenation of several adaptive FBGs. System designs with concatenated adaptive FBGs are outside the scope of this work.
3. Pulse compression in hollow fibers
To further exemplify the functionality of the present system architecture we also set-up an experiment for additional pulse compression in a single-ring HC-PCF [5,6]. As shown in Fig. 1, the hollow fiber is inserted directly after the output of the bulk grating compressor. We used a 30 cm long fiber with a core diameter of 30 µm surrounded by 8 capillaries with a wall thickness of ∼220 nm. The hollow fiber was placed in a pressure chamber equipped with coated windows and after being evacuated was filled with 265 psi of Ar. We obtained a coupling efficiency of 65% from the grating compressor to the hollow fiber, limited by some astigmatism induced by the grating compressor. A dispersive mirror pair was inserted after the hollow fiber to compensate for the dispersion of the output window of the pressure chamber. The temporal profile, FROG traces, spectrum, and phase of the compressed pulses with an output pulse energy of 2.6 µJ at a repetition rate of 2 MHz are shown in Fig. 6 and Fig. 7. We measured a pulse width of around 11.7 fs, which was limited by pulse width smearing [20] in the commercial FROG instrument. About 56% of the pulse energy was contained in the short pulse (the input pulse had ∼70% of the energy in the main pulse.) In the absence of pulse width smearing we calculated that the pulse width would be around 9 fs, which corresponds to 2.5 optical cycles. Even shorter pulses and with higher efficiency can be expected by improving the astigmatism of the compressor and thus improving the coupling into the fiber, as demonstrated by Schade et al. [6]. Carrier phase control of IMRA Yb oscillators is fairly routine and can be implemented with a second Yb fiber amplifier branch, as demonstrated in C. Zhang et al. [1], where a carrier phase noise of <120 mrad was demonstrated. Any inter-branch noise can in principle be compensated via noise cancellation as recently demonstrated by Peng et al., where a relative stability between amplifier branches of ≈ 2 × 10−18 in 1 s was obtained [21].
Fig. 7. For the 11.7 fs pulse: a shows the measured FROG trace, b shows the retrieved FROG trace, and c shows the spectrum in red and the phase in blue.
4. Conclusions
In conclusion we presented a new configuration for producing bandwidth-limited pulses from high energy Yb fiber chirped pulse amplification systems. The inclusion of a single adaptively controlled fiber Bragg grating stretcher is sufficient to compensate for phase distortions due to varying levels of self-phase modulation in the fiber amplifiers, which allows for controllable output pulse energies without any external attenuators while ensuring bandwidth-limited output. Moreover, adjustment of the thermal profile in the fiber Bragg grating stretcher enables the generation of complex pulse shapes such as pulse trains, which can be dialed in via a look-up table or a neural network. The insertion of a single stage Lyot filter maximizes the amplified pulse bandwidth, while efficient pulse compression to the few-cycle pulse regime is obtained with a gas filled HC-PCF. The system can be scaled to significantly higher pulse energies than the demonstrated 10 µJ here via the implementation of longer fiber Bragg grating stretchers and coherent addition. The system can be assembled with all-fiber or micro-optics components apart from a simple bulk grating compressor and is bound to serve as a paradigm for the further dissemination of Yb fiber chirped pulse amplification schemes in science and technology.
Disclosures
The authors declare no conflicts of interest related to this article.
Data availability
Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.
References
1. C. Zhang, P. Li, J. Jiang, L. von der Wense, J. F. Doyle, M. E. Fermann, and J. Ye, “Tunable VUV frequency comb for 229mTh nuclear spectroscopy,” Opt. Lett. 47(21), 5591 (2022). [CrossRef]
2. M. Müller, C. Aleshire, A. Klenke, E. Haddad, F. Légaré, A. Tünnermann, and J. Limpert, “10.4 kW coherently combined ultrafast fiber laser,” Opt. Lett. 45(11), 3083 (2020). [CrossRef]
3. H. Stark, J. Buldt, M. Müller, A. Klenke, and J. Limpert, “100 fs pulses directly from a kW-class mJ-level ytterbium-doped fiber CPA laser system,” Proc. SPIE 11665, Fiber Lasers XVIII: Technology and Systems116651 G (2021). [CrossRef]
4. M. E. Fermann, V. I. Kruglov, B. C. Thomsen, J. M. Dudley, and J. D. Harvey, “Self-Similar Propagation and Amplification of Parabolic Pulses in Optical Fibers,” Phys. Rev. Lett. 84(26), 6010–6013 (2000). [CrossRef]
5. F. Köttig, D. Schade, J. R. Koehler, P. S. J. Russell, and F. Tani, “Efficient single-cycle pulse compression of an ytterbium fiber laser at 10 MHz repetition rate,” Opt. Express 28(7), 9099 (2020). [CrossRef]
6. D. Schade, F. Köttig, J. R. Koehler, M. H. Frosz, P. St, J. Russell, and F. Tani, “Scaling rules for high quality soliton self-compression in hollow-core fibers,” Opt. Express 29(12), 19147 (2021). [CrossRef]
7. A.-K. Raab, M. Seidel, C. Guo, I. Sytcevich, G. Arisholm, A. L’Huillier, C. L. Arnold, and A.-L. Viotti, “Multi-gigawatt peak power post-compression in a bulk multi-pass cell at a high repetition rate,” Opt. Lett. 47(19), 5084 (2022). [CrossRef]
8. P. S. J. Russell, P. Hölzer, W. Chang, A. Abdolvand, and J. C. Travers, “Hollow-core photonic crystal fibres for gas-based nonlinear optics,” Nat. Photonics 8(4), 278–286 (2014). [CrossRef]
9. F. Belli, A. Abdolvand, J. C. Travers, P. St, and J. Russell, “Highly efficient deep UV generation by four-wave mixing in gas-filled hollow-core photonic crystal fiber,” Opt. Lett. 44(22), 5509 (2019). [CrossRef]
10. J. C. Travers, T. F. Grigorova, C. Brahms, and F. Belli, “High-energy pulse self-compression and ultraviolet generation through soliton dynamics in hollow capillary fibres,” Nat. Photonics 13(8), 547–554 (2019). [CrossRef]
11. M. E. Fermann, G Imeshev, I. Hartl, and D. J. Harter, In-line high energy fiber chirped pulse amplification system, U.S. patent # 7,257,302 (30 June 2003).
12. J. Želudevičius, R. Danilevičius, and K. Regelskis, “Optimization of pulse compression in a fiber chirped pulse amplification system by adjusting dispersion parameters of a temperature-tuned chirped fiber Bragg grating stretcher,” J. Opt. Soc. Am. B 32(5), 812 (2015). [CrossRef]
13. K. F. Lee, G. Zhou, Y. Yun, J. Jiang, H. G. Winful, and M. E. Fermann, “Surpassing soliton compression limits in anomalous dispersion high-power erbium fiber comb,” Optica 8(6), 846 (2021). [CrossRef]
14. C. Kerse, H. Kalaycıoğlu, P. Elahi, B. Çetin, D. K. Kesim, Ö. Akçaalan, S. Yavaş, M. D. Aşık, B. Öktem, H. Hoogland, R. Holzwarth, and FÖ Ilday, “Ablation-cooled material removal with ultrafast bursts of pulses,” Nature 537(7618), 84–88 (2016). [CrossRef]
15. M. E. Fermann, F. Haberl, M. Hofer, and H. Hochreiter, “Nonlinear amplifying loop mirror,” Opt. Lett. 15(13), 752 (1990). [CrossRef]
16. M. E. Fermann, M. Hofer, F. Haberl, A. J. Schmidt, and L. Turi, “Additive-pulse-compression mode locking of a neodymium fiber laser,” Opt. Lett. 16(4), 244 (1991). [CrossRef]
17. L. Dong, X. Peng, and J. Li, “Leakage channel optical fibers with large effective area,” J. Opt. Soc. Am. B 24(8), 1689 (2007). [CrossRef]
18. K. F. Lee, A. Rolland, P. Li, J. Jiang, and M. E. Fermann, “Coherent supercontinuum shaping for multiple wavelength optimization over an octave,” Opt. Express 30(1), 427 (2022). [CrossRef]
19. D. N. Schimpf, E. Seise, J. Limpert, and A. Tünnermann, “The impact of spectral modulations on the contrast of pulses of nonlinear chirped-pulse amplification systems,” Opt. Express 16(14), 10664 (2008). [CrossRef]
20. A. Baltuska, M. S. Pshenichnikov, and D. A. Wiersma, “Second-harmonic generation frequency-resolved optical gating in the single-cycle regime,” IEEE J. Quantum Electron. 35(4), 459–478 (1999). [CrossRef]
21. P. Li, A. Rolland, J. Jiang, and M. E. Fermann, “Coherent frequency transfer at < 5*10-21 relative stability via a multi-branch comb with noise cancellation,” Opt. Express 30(13), 22957 (2022). [CrossRef]
