Energy scaling of femtosecond fiber lasers has been constrained by nonlinear impairments and optical fiber damage. Reducing the optical irradiance inside the fiber by increasing mode size lowers these effects. Using an erbium-doped higher-order mode fiber with 6000 µm2 effective area and output fundamental mode re-conversion, we show a breakthrough in pulse energy from a monolithic fiber chirped pulse amplification system using higher-order mode propagation generating 300 µJ pulses with duration <500 fs (FWHM) and peak power >600 MW at 1.55 µm. The erbium-doped HOM fiber has both a record large effective mode area and excellent mode stability, even when coiled to reasonable diameter. This demonstration proves efficacy of a new path for high energy monolithic fiber-optic femtosecond laser systems.
© 2013 Optical Society of America
Generating femtosecond laser pulses  with monolithic fiber-optic amplifier systems provides the ideal form factor for industrial micromachining without a heat affected zone (HAZ) . Energy scaling of femtosecond fiber lasers has been constrained by nonlinear impairments  and optical fiber damage. Reducing the optical irradiance inside the fiber by increasing mode size lowers these effects. However, all established approaches [4–13] propagate light in the fundamental mode (LP0,1), where both bend-induced reduction in mode size and mode instability [11,14] are more pronounced. The energy scaling of femtosecond chirped pulse amplification (CPA)  fiber laser systems is limited by temporal pulse distortion caused by nonlinear effects. The dominant limiter is self-phase modulation (SPM)  as the pulses propagate through the system amplifiers. The magnitude of SPM is quantified by the total nonlinear phase shift [17,18], so called B-integral, which is proportional to the optical irradiance inside the fiber.
For a given wavelength, material (with nonlinear index n2), pulse peak power, and fiber length, the only way to reduce the accumulated nonlinearity is by increasing the fiber effective mode area Aeff. Improved fibers allow a reduced optical irradiance at high power by increasing Aeff. However, to achieve Aeff much greater than ~300 µm2, one must overcome severe fundamental and practical challenges. To maintain diffraction-limited spatial beam quality, these large mode area (LMA) fibers have to be designed and utilized so as to ensure robust single-mode propagation. Conventional single-mode fibers can, in theory, be adapted to provide a very large Aeff by increasing core diameter and lowering numerical aperture (NA). However, doing so in practice results in a very weak waveguide, and the optical fiber becomes very sensitive to mechanical disturbance. To date, most efforts to develop LMA fibers have focused on approaches such as low NA photonic crystal fiber (PCF) [4–8], leakage channel fiber (LCF) , chirally-coupled core fiber (CCC) , and large-pitch fiber . These all operate in the fundamental mode (LP0,1) and are prone to significant bend-induced reduction in mode size and mode instability (energy coupling to multiple modes). These approaches can only be scaled to Aeff >1000 µm2 if the fibers are kept straight to avoid mode distortion and multimode propagation . Thus fibers have been formed as rigid rods with a diameter of 1.5 mm or more [6,20]. These rods however are limited to shorter lengths (around 1 m) due to practical packaging constraints. In turn, this can significantly limit the available gain, energy storage, and average power of the laser system. In addition, due to its large diameter, it is very challenging to splice such rods to fibers or fiber components. Consequently, progress towards bendable fibers with large Aeff has been stagnant in recent years. Nonetheless, the overwhelming size, weight, and power advantages of monolithic fiber-optic femtosecond laser systems provoke compelling motivation to scale their pulse energy. In a break from conventional thinking, we achieve high energy, excellent pulse and beam quality, and sufficient thermal management using erbium (Er)-doped higher-order mode (HOM) fiber as the main amplifier medium in our CPA system. In this paper, we demonstrate for the first time a monolithic fiber-optic CPA system generating near diffraction limited, fully compressed pulses with 300 µJ energy, <500 fs duration (FWHM), and >600 MW peak power at 1.55 µm wavelength.
2. HOM fiber amplifier for high energy CPA systems
HOM amplifiers have recently proven highly effective for scalable mode size, resistance to bend-induced mode size reductions, and superior stability [11,12]. The large mode size of the HOM also provides high saturation energy required for high energy pulse amplification. Er-doped HOM amplifiers can be core-pumped with both pump and signal propagating in the same mode, allowing for maximum pump-signal overlap, high pump absorption, and reduced amplifier length (lower accumulated nonlinearity).
The final stage of amplification is performed in an Er-doped HOM fiber with Aeff = 6000 µm2 in LP0,11 mode, as shown in Fig. 1. The Er density is maximized to provide 50 dB/m in-band absorption and optimal amplifier length of 2.65 m. For these experiments, the fiber was coiled to radius of only 20 cm. The area of the LP0,1 mode in this fiber is 110 µm2 (vs. 75 µm2 for standard single mode fiber). The input of the Er-doped HOM fiber has a long period grating (LPG) to convert the LP0,1 mode of both the 1480 nm pump and the 1553 nm signal to the LP0,11 mode. Figure 1 (top left) shows the LP0,1 mode transmission of the LPG vs. wavelength. Because the strong Er absorption band is within the LPG conversion band, it was difficult to measure the optical spectrum of the LPG through long lengths of the Er-doped HOM fiber. A typical LPG transmission was measured with shorter Er-doped HOM fiber during LPG development . The 20 dB (99%) conversion efficiency bandwidth is >100 nm covering both the pump wavelength (1480 nm) and the signal wavelength (1553 nm).
An identical LPG near the end of the HOM fiber re-converts the amplified signal from the LP0,11 mode back to LP0,1 mode with Aeff = 110 µm2. The HOM fiber was terminated 5 mm after the re-converting LPG and end-capped with a 750 µm length of 330 µm diameter coreless fiber, angle polished at 8°. The entire Er-doped HOM fiber amplifier assembly is mounted on a 22 °C water chilled aluminum plate. This new HOM fiber amplifier generates pulse energy 3 × higher than recently reported monolithic fiber femtosecond laser demonstrations  that feature low M2 and fiber format that can be fusion spliced and coiled to reasonable bend radii for packaging.
3. Experiment and results
The novel HOM fiber amplifier is tested in a monolithic fiber CPA circuit to assess its performance and relevance to industrial laser systems. Figure 2 shows a simplified schematic of the amplifier test bed CPA architecture used to minimize SPM during the amplification of high peak power pulses (>100 MW) . The all-fiber seed laser comprises these elements: mode-locked laser, spectral broadening amplifier, fiber pulse stretcher, active pulse shaper, pre-amplifier #1, AOM (picker #1), and a pre-amplifier #2. The seed source provides a signal pulse train with 1.7 ns (FWHM) stretched pulse duration, 25 kHz pulse repetition frequency, and 9.3 mW average power. The 1480 nm Raman fiber pump laser has maximum power of 70 W. Pump and 1553 nm signal are combined by a fused single mode, high power wavelength division multiplexer (WDM).
Briefly, pulses from a mode-locked laser (MLL) are spectrally broadened via controlled SPM in a low-gain spectral broadening amplifier, stretched to ~1.7 ns by a pair of chirped fiber Bragg gratings (CFBG), amplified through a sequence of erbium fiber amplifiers, and compressed to <500 fs in a Treacy style pulse compressor . Pulse rate is reduced from 40 MHz down to 25 kHz between pre-amplifiers using a >35 dB extinction acousto-optic modulator (AOM), referred to in the graphic as picker #1. To further extend the pulse energy headroom via pre-compensation for phase distortion , we implement a commercial spectral processor (Finisar 1000E) based on liquid crystal on silicon (LCoS) technology as an active pulse shaper between pre-amplifiers. These light generation and amplification components are all fiber optic and fusion spliced together to form a stable, and rugged optical path. The evolution of the pulse duration and its relative intensity along the monolithic fiber-optical chain and free-space compressor is notionally illustrated in Fig. 2 (above schematic). Final output control through a second AOM (picker #2) and the pulse compression are performed in an industrialized free-space opto-mechanical assembly.
A band-pass filter deflects residual 1480 nm pump light at the amplifier output. A quarter wave plate and half wave plate pair and an optical isolator respectively optimize the signal state of polarization and block back-reflections. The incidence angle and inter-grating distance are optimized in the Treacy compressor to de-chirp/recompress the pulses with minimal pulse duration and pulse pedestal.
The booster output signal power is measured over a range of pump power. The pulse energy is calculated by dividing the measured signal optical power by the pulse repetition frequency. The signal optical spectra are recorded with an optical spectrum analyzer (YOKOGAWA, 735301). At the booster amplifier and compressor output, a small fraction of the beam is sampled using a pellicle beam splitter to assess spatial beam quality (M2) and temporal pulse quality. Spatial beam measurements are performed with a scanning slit beam profiler and a near infrared (NIR) CCD camera (SPIRICON, SP503U-1550) near focus of a 400 mm focal length lens. Pulse temporal quality is measured using a common second harmonic generation (SHG) intensity autocorrelator (FEMTOCHROME RESEARCH, FR-103HS-XL) and compared to a reference mathematical pulse profile to estimate experimental pulse shape and duration.
Figure 3(a) demonstrate scaling up to maximum pulse energy of 595 µJ and average power of 14.87 W (gain = 32 dB), free of nonlinear limits. The HOM amplifier’s output signal pulse energy and average power are plotted vs. 1480 nm pump power at 25 kHz pulse repetition frequency and 9.3 mW input signal power. It should be noted that there is no pump-to-signal roll-over, confirming that the large mode area fiber overcomes thermal and nonlinear optical induced efficiency limits. The efficiency derived from the linear curve fit shown in Fig. 3(a) is 25%. The optical spectra of the input signal and the output signal at pulse energy ranging from 16 µJ to 595 µJ are shown in Fig. 3(b). The signal power near 1553 nm center wavelength is more than 20 dB higher than the background—comprised of amplified spontaneous emission (ASE) power near 1530 nm and long wavelength (>1562 nm) spectral broadening. Even for the highest pulse energy, the signal-to-noise (SNR) ratio is greater than 20 dB. Together, these performance features confirm the amplifier’s excellent gain dynamics and low nonlinear phase accumulation. The estimated B-integral for the Er-doped HOM amplifier is 2π radians, with about 0.5π radians generated during propagation in the 5 mm span of LP0,1 mode after the re-converting LPG.
In addition to basic performance as an optical amplifier, we characterized the output signal temporal and spatial quality. Using the optical spectrum measured at 595 µJ (amplifier output), the transform limited pulse FWHM is 370 fs. The FWHM of the measured autocorrelation trace is 717 fs. We have used a sech2 pulse autocorrelation with assumed 0.65 deconvolution factor to estimate the duration of the pulse to be 466 fs, ~1.26x the transform limited pulse FWHM. The maximum pulse energy at the compressor output is 300 µJ, with 50% compressor efficiency. The estimated peak power of these pulses is ~644 MW. The actual pulse duration may be slightly longer than 466 fs considering the difference between the actual pulse shape and the assumed sech2 shape. Similarly, the actual peak power may be lower than the estimated peak power due to its non sech2 shaped pulse and portion of the energy that lies in the pedestal. The theoretical sech2-shaped pulse autocorrelation is included in Fig. 4(a) for reference (red dashed line). The measured beam propagation parameter, or M2, at the maximum pulse energy of 595 µJ at the amplifier output was 1.51 (horizontal) and 1.21 (vertical), demonstrating nearly diffraction limited spatial beam quality. We believe further optimization of both the input and the output LPGs will further improve the spatial beam quality. The inset of Fig. 4(b) shows the transverse beam profile measured near the focus of a reference lens (f = 400 mm plano-convex singlet).
Utilizing novel HOM fiber with record large mode area in a CPA laser system, we have for the first time demonstrated the capability to produce fully compressed pulses with near diffraction limited beam and near transform limited pulses with record femtosecond pulse energy of 300 μJ at 1.55 µm center wavelength. This pulse energy demonstration represents an extraordinary advancement over previous fiber CPA systems and proves that robust monolithic fiber laser architecture can support much higher energy scaling for industrial femtosecond laser applications. Increases in Aeff of the HOM fiber will proportionally raise the maximum pulse energy at a given B-integral without imposing significant bend-induced reduction in the mode size or increase in modal instability. In addition, when coupled with developments to further manage nonlinearities, such as phase tailoring and enhanced pulse chirping ratios [24,25], the maximum pulse energy of monolithic, fiber-optic femtosecond laser systems could reach multiple millijoule levels in the near future.
References and links
1. P. M. W. French, “The generation of ultrashort laser pulses,” Rep. Prog. Phys. 58(2), 169–262 (1995). [CrossRef]
2. F. Abedin and D. K. Kalla, “Review on heat affected zone (HAZ) in laser machining,” in Proceedings of the 6th Annual GRASP Symposium (Wichita State University, 2010), pp. 63–64.
3. S. P. Singh and N. Singh, “Nonlinear effects in optical fibers: origin, management and applications,” Prog. Electromagn. Res. 73, 249–275 (2007). [CrossRef]
6. J. Limpert, N. Deguil-Robin, I. Manek-Hönninger, F. Salin, F. Röser, A. Liem, T. Schreiber, S. Nolte, H. Zellmer, A. Tünnermann, J. Broeng, A. Petersson, and C. Jakobsen, “High-power rod-type photonic crystal fiber laser,” Opt. Express 13(4), 1055–1058 (2005). [CrossRef] [PubMed]
7. J. C. Knight, T. A. Birks, R. F. Cregan, P. S. J. Russell, and J. P. de Sandro, “Large mode area photonic crystal fiber,” Electron. Lett. 34(13), 1347–1348 (1998). [CrossRef]
8. K. Hougaard and F. D. Nielsen, “Amplifiers and lasers in PCF configurations,” J. Opt. Fiber Commun. 1(1), 63–83 (2004). [CrossRef]
9. L. Dong, X. Peng, and J. Li, “Leakage channel optical fibers with large effective area,” J. Opt. Soc. Am. B 24(8), 1689–1697 (2007). [CrossRef]
10. J. Želudevičius, R. Danilevičius, K. Viskontas, N. Rusteika, and K. Regelskis, “Femtosecond fiber CPA system based on picosecond master oscillator and power amplifier with CCC fiber,” Opt. Express 21(5), 5338–5345 (2013). [CrossRef] [PubMed]
12. S. Ramachandran, J. W. Nicholson, S. Ghalmi, M. F. Yan, P. Wisk, E. Monberg, and F. V. Dimarcello, “Light propagation with ultralarge modal areas in optical fibers,” Opt. Lett. 31(12), 1797–1799 (2006). [CrossRef] [PubMed]
13. J. Limpert, F. Stutzki, F. Jansen, H. J. Otto, T. Eidam, C. Jauregui, and A. Tünnermann, “Yb-doped large-pitch fibres: effective single-mode operation based on higher-order mode delocalisation,” Light Sci. Appl. 1(4), 1–5 (2012).
14. J. W. Nicholson, J. M. Fini, A. M. DeSantolo, X. Liu, K. Feder, P. S. Westbrook, V. R. Supradeepa, E. Monberg, F. DiMarcello, R. Ortiz, C. Headley, and D. J. DiGiovanni, “Scaling the effective area of higher-order-mode erbium-doped fiber amplifiers,” Opt. Express 20(22), 24575–24584 (2012). [CrossRef] [PubMed]
15. M. Pessot, P. Maine, and G. Mourou, “1000 times expansion/compression of optical pulses for chirped pulse amplification,” Opt. Commun. 62(6), 419–421 (1987). [CrossRef]
17. J. Limpert, F. Roser, T. Schreiber, and A. Tunnermann, “High-power ultrafast fiber laser systems,” IEEE J. Sel. Top. Quantum Electron. 12(2), 233–244 (2006). [CrossRef]
18. A. Galvanauskas, “Ultrashort-pulse fiber amplifiers,” in Ultrafast Lasers: Technology and Applications, M. Fermann, A. Galvanauskas, and G. Sucha, eds. (Marcel Dekker, 2002), pp. 155–218.
19. J. Li, X. Peng, and L. Dong, “Robust fundamental mode operation in a ytterbium-doped leakage channel fiber with an effective area of ~3000μm2,” in Advanced Solid-State Photonics, OSA Technical Digest Series (CD) (Optical Society of America, 2007), paper ME3.
20. T. Eidam, J. Rothhardt, F. Stutzki, F. Jansen, S. Hädrich, H. Carstens, C. Jauregui, J. Limpert, and A. Tünnermann, “Fiber chirped-pulse amplification system emitting 3.8 GW peak power,” Opt. Express 19(1), 255–260 (2011). [CrossRef] [PubMed]
21. X. Peng, K. Kim, M. Mielke, S. Jennings, G. Masor, D. Stohl, A. Chavez-Pirson, D. T. Nguyen, D. Rhonehouse, J. Zong, D. Churin, and N. Peyghambarian, “High efficiency, monolithic fiber chirped pulse amplification system for high energy femtosecond pulse generation,” Opt. Express 21(21), 25440–25451 (2013). [CrossRef] [PubMed]
22. E. Treacy, “Optical pulse compression with diffraction gratings,” IEEE J. Quantum Electron. 5(9), 454–458 (1969). [CrossRef]
23. D. Nguyen, M. U. Piracha, and P. J. Delfyett, “Transform-limited pulses for chirped-pulse amplification systems utilizing an active feedback pulse shaping technique enabling five time increase in peak power,” Opt. Lett. 37(23), 4913–4915 (2012). [CrossRef] [PubMed]
24. J. Limpert, F. Roser, D. N. Schimpf, E. Seise, T. Eidam, S. Hadrich, J. Rothhardt, C. J. Misas, and A. Tunnermann, “High repetition rate gigawatt peak power fiber laser systems: challenges, design, and experiment,” IEEE J. Sel. Top. Quantum Electron. 15(1), 159–169 (2009). [CrossRef]
25. L. Shah, Z. Liu, I. Hartl, G. Imeshev, G. C. Cho, and M. E. Fermann, “Ultrafast high energy amplifiers beyond the B-integral limit,” Proc. SPIE 6102, 61020Z (2006). [CrossRef]