Next Article in Journal
A Motion Compensation Method for Terahertz SAR Imaging with a Large Squint
Next Article in Special Issue
Wavelength Conversion Process of Intra-Pulse Stimulated Raman Scattering in Near-Zero Negative Dispersion Range
Previous Article in Journal
Broadening the Bandwidth of Epsilon-Near-Zero Metamaterials with Embedded Square Frames
Previous Article in Special Issue
Random Raman Lasing in Diode-Pumped Multi-Mode Graded-Index Fiber with Femtosecond Laser-Inscribed Random Refractive Index Structures of Various Shapes
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Communication

Excess Intensity Noise in a Nonlinear Amplifying Loop-Mirror-Based Mode-Locked Laser from a Non-Reciprocal Phase Bias

1
Korea Research Institute of Standards and Science (KRISS), Daejeon 34113, Republic of Korea
2
Precision Measurement, University of Science and Technology (UST), Daejeon 34113, Republic of Korea
Photonics 2024, 11(12), 1186; https://doi.org/10.3390/photonics11121186
Submission received: 20 November 2024 / Revised: 11 December 2024 / Accepted: 16 December 2024 / Published: 18 December 2024
(This article belongs to the Special Issue Advancements in Fiber Lasers and Their Applications)

Abstract

:
We demonstrate a low-intensity-noise, nonlinear amplifying loop-mirror-based mode-locked fiber laser by optimizing the polarization of the non-reciprocal phase bias and the pump current. If the angle of the waveplate in the non-reciprocal phase bias to the polarization axis of a polarization-maintaining fiber is not carefully aligned, parasitic polarization is induced. The parasitic polarization affects the minimum pump power and dynamic range of pump power for mode-locking, the intensity noise, and the comb power. To reduce intensity noise, the angle of the waveplate for the non-reciprocal phase bias is adjusted, and then the pump power is adjusted. The waveplate angle minimizing the intensity noise maximizes the dynamic range of the pump power for mode-locking and output power. As a result, the relative intensity noise has been suppressed by more than 32 dB at 15 kHz Fourier frequency. The polarization extinction ratio at the non-reciprocal phase bias is critical since it can determine a cavity loss and quality factor of a laser oscillator. Therefore, the additional polarizers cannot improve the intensity noise once the angle is mismatched and the polarization extinction ratio is degraded.

1. Introduction

Femtosecond mode-locked lasers, often so-called optical frequency comb when fully-locked, have been used in various fields from fundamental science to high precision industrial applications, known as time and frequency measurement [1] and transfer of optical clocks [2], timing synchronization of X-ray [3] and free-electron laser systems [4], ultralow phase-noise microwave generation [5], dimensional metrology such as photonic radar [6], ranging [7,8,9], surface and transient response measurement [10], astrophysical measurement [11,12], very long baseline interferometry [13], and comb-based spectroscopy [14,15,16].
In particular, femtosecond mode-locked lasers are very powerful tools in spectroscopy since they can cover a broad spectrum, have high-speed sampling of more than megahertz, and are coherent that can be transferred in free space [14]. In the last two decades, one of the powerful applications of comb-based spectroscopy is dual-comb spectroscopy. Dual-comb spectroscopy system can cover the broadband which is the advantage of Fourier transform infrared interferometry, while it does not require any mechanical moving part to get interferogram by detuning the repetition rate of two combs. At the same time, the sampling speed is typically half of the repetition rate, which is faster than microsecond timescale that can measure the molecular reaction in real time.
More recently, mid-infrared (mid-IR) generation has been enabled by a single comb source [17,18] supported by a process known as intrapulse difference frequency generation (DFG) without an optical parametric oscillation (OPO) process that can reduce the complexity of the mid-IR generation process. Therefore, mid-IR dual-comb spectroscopy is going to be a more powerful tool for greenhouse gas research and biomedical research [19].
Although intrapulse DFG-based mid-IR generation is getting simpler, it is still based on a nonlinear process that requires a stable pulse train in terms of pulsewidth and pulse power. It needs a stable and short pulse, but a short pulse is vulnerable to pulse broadening due to the unwanted nonlinear process in fiber during amplification and transfer. Therefore, minimizing the intrinsic intensity noise of the seed source is highly beneficial. It is well known that the mode-locked laser with an artifcial saturable absorber, such as nonlinear polarization rotation and nonlinear amplifying loop mirror, can achieve low-intensity noise.
In this paper, we studied the excess intensity noise from the polarization status of non-reciprocal phase bias in a nonlinear amplifying loop-mirror-based mode-locked laser. The parasitic polarization in a non-reciprocal phase bias can degrade the intensity noise by more than 32 dB. Since the parasitic polarization degrades the quality factor of a mode-locked laser, the extra polarizer inside a laser oscillator or after the output of a laser does not help to suppress the excess intensity noise.

2. Methods

A home-built, polarization maintaining (PM), nonlinear amplifying loop-mirror-based (NALM) Erbium mode-locked laser with a repetition rate of 207.3 MHz is used in this experiment. High repetition rate femtosecond lasers are advantageous in getting a high mode-power in the frequency domain for spectroscopy. In the laser oscillator, a 40-cm-long Er-doped gain fiber (110 dB/m PM fiber) is spliced to a PM collimator, and a PM wavelength division multiplexer (WDM), and it is pumped by a PM pump diode (950 mW, 976 nm). The coupling ratio of the PM fiber coupler is 60:40. Two ports of the PM coupler are coupled to PM fibers and the other two ports of the PM coupler are coupled to free space. A dielectric mirror is attached to a piezoelectric transducer for the linear cavity of a nonlinear amplifying loop mirror. The output of the PBS in the fiber loop is used for the experiment after a PM isolator.
In order to characterize the intensity noise sensitivity from non-reciprocal phase bias, the angle of the quarter waveplate in the non-reciprocal phase bias is adjusted ~40°. The non-reciprocal phase bias adjusts the phase bias between the clockwise pulse train and the counter-clockwise pulse train that can change the transmission of the fiber loop (i.e., Sagnac loop) in a nonlinear amplifying loop-mirror. Therefore, it works as an artificial saturable absorber of a mode-locked laser since the higher power light corresponds to the high transmittance while the lower power light corresponds to the low transmittance that can shape the pulse. The output after a fiber-coupled isolator is used and measured in this experiment. The output power is typically more than 1 mW.
The RF spectra of harmonics of repetition rate up to 10 GHz in Figure 1b are measured with a 26.5 GHz-bandwidth photodetector. The RF spectrum of a fundamental repetition rate with a 10 MHz span in Figure 1c shows no spur or bump, which is enabled by stable mode-locking. The 3-dB bandwidth of the optical spectrum is 24 nm centered at 1565 nm, as shown in Figure 1d. The typical relative intensity noise of the demonstrated laser is shown in Figure 1e. The relative intensity noise (RIN) is measured by a 100 MHz bandwidth high-impedance photodetector and analyzed by a FFT analyzer (<100 kHz Fourier frequency) and an RF analyzer (>100 kHz Fourier frequency).

3. Results and Discussion

Many factors determine the intensity noise of polarization-maintaining (PM) nonlinear amplifying loop-mirror-based mode-locked lasers, such as the pump power [20], output coupling ratio, coupling ratio of the loop mirror, and polarization extinction ratio. This paper primarily examines the polarization optimization of a non-reciprocal phase bias and subsequent pump power optimization to minimize the intensity noise. To study the polarization dynamics under non-reciprocal phase bias, we systematically adjust the angle of the quarter waveplate in the non-reciprocal phase bias shown in Figure 1a. The quarter waveplate is employed in our system due to its advantage of facilitating a π/2 phase bias, which maximizes the differential transmission of light propagating in opposite directions within the fiber loop. The intensity noise is measured with a 10°-step. The pump power is accordingly adjusted at each step to achieve the minimum intensity noise condition for each angle. The lowest intensity noise of this system is labeled as 0°.
As shown in Figure 2a, the intensity noise difference is observed at Fourier frequencies below 100 kHz. The broadband peaks below 500 Hz are mainly from the acoustic noise in the lab. The noise floor above 100 kHz is the photodetection shot noise. The comb power is above 1 mW, and the corresponding shot noise is −156 dB/Hz. However, we split the output power for the intensity noise and optical spectrum measurement. The typical optical power for RIN measurement is about 300 μW, corresponding to about −150 dB/Hz. The maximum RIN difference is observed between 0° and −20° (curve (i) and curve (iii) in Figure 2a. The maximum difference between the curves is 32 dB at 15 kHz Fourier frequency. The main reason is the parasitic circular polarization due to the angle mismatch of the quarter waveplate in the phase bias to the slow axis of PM fibers in the laser oscillator. The quarter waveplate inside the non-reciprocal phase bias employs the exact π/2 phase bias between the traveling light in the fiber loop in opposite directions. However, this only holds when the traveling light in opposite directions is linearly polarized, and it has to travel the fast and slow axis of the quarter waveplate, respectively. The parasitic polarization reduces the effective gain because it splits the power to each axis and, as a result, reduces the nonlinear phase bias between opposite directions. The reduced phase shift deteriorates the pulse shaping in a nonlinear amplifying loop mirror and affects the pulse stability (i.e., intensity fluctuation). Unfortunately, it is difficult to measure the polarization extinction ratio (PER) with respect to each angle because most of the components in this laser oscillator work on only a slow axis (i.e., it works as a polarizer).
The angle dependence due to the parasitic polarization is also observed in the optical spectra, as shown in Figure 2b. As the angle changes, the power of the sideband (e.g., 1520 nm and 1620 nm) changes. As the angle deviates from the best point, the spectrum gets narrow while the center wavelength remains the same. However, the intensity noise or optical spectrum is not symmetric with respect to 0°. One possible reason is that the waveplate and PM Er fiber are not on the axis. Another reason is the backlash of the waveplate.
The lowest intensity noise of each QWP position is achieved by adjusting the pump power. The mode-locking maintains more than 45° variation of the QWP angle. As mentioned above, the curves denoted by 0° mean the QWP axis and polarization do not generate the parasitic polarization that can achieve the lowest intensity noise condition. For the lowest intensity noise condition, the working range of the pump power that can maintain the mode-locking is more than any other angle. If the pump power decreases below the lowest value for each QWP angle plotted in Figure 3, the mode-locking is not available. If the pump power increases more than the highest value plotted in Figure 3, the generated continuous-wave (CW) peak makes the mode-lock unstable, resulting in intensity fluctuation. The Erbium gain fiber is not long enough to absorb the pump power because the laser oscillator targets the high repetition rate. Therefore, the remaining CW light is not beneficial. The pump power of each QWP angle for the lowest intensity noise condition in Figure 2 is 355 mW (Figure 3(i)), 530 mW (Figure 3(ii), Figure 3(iv), Figure 3(v)), and 588 mW (Figure 3(iii)), respectively.
The laser oscillator in this system is based on a polarization-maintaining fiber, so it is easy to make the work only on a slow-axis by using only slow-axis working fiber components. Therefore, the polarization extinction ratio (PER) measured at the output before the isolator (working in both axis) in Figure 1a is 19 dB, regardless of the QWP angle in the non-reciprocal phase bias. However, the polarization extinction ratio at the PM fiber is not as high as the polarization beam splitter (PBS), the excess or residual unpolarized light can affect the intensity noise performance. To verify the difference, the additional PBSs are added to input and output of the non-reciprocal phase bias but the effect is negligible. At last, the additional polarizer is added to the output because it is the last stage to measure the intensity noise. Although PER increases to 25 dB (the same as in the spec sheet), the intensity noise is not improved at all, as shown in Figure 4.

4. Conclusions

The intensity noise of a nonlinear-amplifying loop mirror-based mode-locked laser is studied by adjusting the waveplate in a non-reciprocal phase bias. Since the waveplate in a non-reciprocal phase bias has a significant role in a nonlinear-amplifying loop mirror-based mode-locked laser, the angle mismatch can degrade the intensity noise by more than 30 dB compared to the best state. The mismatched angle between the waveplate and the slow axis of the PM fiber results in the parasitic polarization. The parasitic polarization reduces the effective gain of the mode-locked laser. Therefore, we observe a narrower optical spectrum, a narrower range of pump power required for mode-locking, and less comb power is generated. Even if a polarization beam splitter or a polarizer is added to reduce the parasitic polarization, the intensity noise is not improved because the quality factor of the laser oscillator is already degraded (i.e., lossy due to the degraded effective gain).
Intensity noise is often considered less important than other noises, such as timing jitter, linewidth, and frequency noise. However, it becomes an important noise factor because it determines the center wavelength of the generated mid-IR source, which gives the frequency traceability of the spectroscopy data. In addition, the intensity noise is coupled to timing jitter, linewidth, and frequency noise by the fiber dispersion. Therefore, it is recommended to optimize the pump power and align the non-reciprocal phase bias to obtain an intrinsically low-intensity noise mode-locked laser.

Funding

This research was funded by the Korea Meteorological Administration Research and Development Program under Grant KMI2022-01410.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

The data presented in this paper are available upon resonable request.

Conflicts of Interest

The author declares no conflicts of interest.

References

  1. Hollberg, L.; Diddams, S.; Bartels, A.; Fortier, T.; Kim, K. The measurement of optical frequencies. Metrologia 2005, 42, 105–124. [Google Scholar] [CrossRef]
  2. Boulder Atomic Clock Optical Network (BACON) Collaboration. Frequency ratio meausrements at 18-digit accuracy using an optical clock network. Nature 2021, 591, 564–569. [Google Scholar] [CrossRef] [PubMed]
  3. Kim, J.; Cox, J.A.; Chen, J.; Kärtner, F.X. Drift-free femtosecond timing synchronization of remote optical and microwave sources. Nat. Photonics 2008, 2, 733–736. [Google Scholar] [CrossRef]
  4. Schulz, S.; Grguraš, I.; Behrens, C.; Bromberger, H.; Costello, J.T.; Czwalinna, M.K.; Felber, M.; Hoffmann, M.C.; Ilchen, M.; Liu, H.Y.; et al. Femtosecond all-optical synchronization of an X-ray free-electron laser. Nat. Commun. 2015, 6, 5938. [Google Scholar] [CrossRef] [PubMed]
  5. Xie, X.; Bouchand, R.; Nicolodi, D.; Guinta, M.; Hänsel, W.; Lezius, M.; Joshi, A.; Datta, S.; Alexandre, C.; Lours, M.; et al. Photonic microwave signals with zeptosecond-level absolute timing noise. Nat. Photonics 2017, 11, 44–47. [Google Scholar] [CrossRef]
  6. Ghelfi, P.; Laghezza, F.; Scotti, F.; Serafino, G.; Capria, A.; Pinna, S.; Onori, D.; Porzi, C.; Scaffardi, M.; Malacarne, A.; et al. A fully photonics-based coherent radar system. Nature 2014, 507, 341–345. [Google Scholar] [CrossRef] [PubMed]
  7. Lee, J.; Kim, Y.; Lee, K.; Lee, S.; Kim, S. Time-of-flight measurement with femtosecond light pulses. Nat. Photonics 2010, 4, 716–720. [Google Scholar] [CrossRef]
  8. Shi, H.; Song, Y.; Liang, F.; Xu, L.; Hu, M.; Wang, C. Effect of timing jitter on time-of-flight distance measurements using dual femtosecond lasers. Opt. Express 2015, 23, 14057–14069. [Google Scholar] [CrossRef] [PubMed]
  9. Ahn, C.; Na, Y.; Kim, J. Dynamic absolute distance measurement with nanometer-precision and MHz acquisition rate using a frequency comb-based combined method. Opt. Lasers Eng. 2023, 162, 107414. [Google Scholar] [CrossRef]
  10. Na, Y.; Kwak, H.; Ahn, C.; Lee, S.; Lee, W.; Kang, C.-S.; Lee, J.; Suh, J.; Yoo, H.; Kim, J. Massively parallel electro-optic sampling of space-encoded optical pulses for ultrafast mul-ti-dimensional imaging. Light: Sci. Appl. 2023, 12, 44. [Google Scholar] [CrossRef] [PubMed]
  11. Obrzud, E.; Rainer, M.; Harutyunyan, A.; Anderson, M.H.; Geiselmann, M.; Chazelas, B.; Kundermann, S.; Lecomte, S.; Cecconi, M.; Ghedina, A.; et al. A microphotonic astrocomb. Nat. Photonics 2019, 13, 31–35. [Google Scholar] [CrossRef]
  12. Probst, R.A.; Milaković, D.; Toledo-Padrón, B.; Lo Curto, G.; Avila, G.; Brucalassi, A.; Holzwarth, R.; Martins, B.L.C.; Esposito, M.; Grupp, F.; et al. A crucial test for astronomical spectro-graph calibration with frequency combs. Nat. Astron. 2020, 4, 603–608. [Google Scholar] [CrossRef]
  13. Clivati, C.; Aiello, R.; Bianco, G.; Bortolotti, C.; Natale, P.D.; Di Sarno, V.; Maddaloni, P.; Maccaferri, G.; Mura, A.; Negusini, M.; et al. Common-clock very long baseline interferometry using a coherent optical fiber link. Optica 2020, 7, 1031–1037. [Google Scholar] [CrossRef]
  14. Coddington, I.; Newbury, N.; Swann, W. Dual-comb spectroscopy. Optica 2016, 3, 414–426. [Google Scholar] [CrossRef] [PubMed]
  15. Bernhardt, B.; Ozawa, A.; Jacquet, P.; Jacquey, M.; Kobayashi, Y.; Udem, T.; Holzwarth, R.; Guelachvili, G.; Hänsch, T.W.; Picqué, N. Cavity-enhanced dual-comb spectroscopy. Nat. Photonics 2010, 4, 55–57. [Google Scholar] [CrossRef]
  16. Adler, F.; Thorpe, M.J.; Cossel, K.C.; Ye, J. Cavity-enhanced direct frequency comb spectroscopy: Technology and applications. Ann. Rev. Anal. Chem. 2010, 3, 175–205. [Google Scholar] [CrossRef] [PubMed]
  17. Lesko, D.M.; Timmers, H.; Xing, S.; Kowligy, A.; Lind, A.J.; Diddams, S.A. A six-octave optical frequency comb from a scala-ble few-cycle erbium fibre laser. Nat. Photonics 2021, 15, 281–286. [Google Scholar] [CrossRef]
  18. Hoghooghi, N.; Xing, S.; Chang, P.; Lesko, D.; Lind, A.; Rieker, G.; Diddams, S. Broadband 1-GHz mid-infrared frequency comb. Light Sci. Appl. 2022, 11, 264. [Google Scholar] [CrossRef] [PubMed]
  19. Chang, P.; Ishrak, R.; Hoghooghi, N.; Egbert, S.; Lesko, D.; Swartz, S.; Biegert, J.; Rieker, G.B.; Reddy, R.; Diddams, S.A. Mid-Infrared Hyperspectral Microscopy with Broadband 1-GHz Dual Frequency Combs. APL Photonics 2024, 9, 106111. [Google Scholar] [CrossRef]
  20. Kwon, D.; Kim, D. Ultralow Intensity Noise Pulse Train from an All-fiber Nonlinear Amplifying Loop Mirror-based Femtosecond Laser. Curr. Opt. Photonics 2023, 7, 708–713. [Google Scholar] [CrossRef]
Figure 1. (a) System description of a 207-MHz, nonlinear amplifying loop-mirror-based mode-locked laser. Er, Erbium-doped gain fiber; WDM, wavelength-division multiplexer, φ, non-reciprocal phase bias; FR, Faraday-rotator; QWP, quarter-waveplate; (b) RF spectrum measurement of repetition-rate with a high-bandwidth photodetector; (c) RF spectrum of a fundamental repetition-rate; (d) Optical spectrum in log-scale (left) and linear scale (right); (e) Relative intensity noise.
Figure 1. (a) System description of a 207-MHz, nonlinear amplifying loop-mirror-based mode-locked laser. Er, Erbium-doped gain fiber; WDM, wavelength-division multiplexer, φ, non-reciprocal phase bias; FR, Faraday-rotator; QWP, quarter-waveplate; (b) RF spectrum measurement of repetition-rate with a high-bandwidth photodetector; (c) RF spectrum of a fundamental repetition-rate; (d) Optical spectrum in log-scale (left) and linear scale (right); (e) Relative intensity noise.
Photonics 11 01186 g001
Figure 2. (a) The lowest relative intensity noise of different QWP angle after pump power optimization; (b) The optical spectra of different QWP angle.
Figure 2. (a) The lowest relative intensity noise of different QWP angle after pump power optimization; (b) The optical spectra of different QWP angle.
Photonics 11 01186 g002
Figure 3. Pump power and corresponding comb power for each QWP angle. (i) −20°, black square, (ii) −10°, red circle, (iii) 0°, green triangle, (iv) 10°, blue star, (v) 20°, violet pentagon.
Figure 3. Pump power and corresponding comb power for each QWP angle. (i) −20°, black square, (ii) −10°, red circle, (iii) 0°, green triangle, (iv) 10°, blue star, (v) 20°, violet pentagon.
Photonics 11 01186 g003
Figure 4. Relative intensity noise measurement when an additional polarizer is inserted at the output (denoted as w/polarizer). Curves (i) and (ii), curves (iii) and (iv), curves (v) and (vi) are operated at the same condition.
Figure 4. Relative intensity noise measurement when an additional polarizer is inserted at the output (denoted as w/polarizer). Curves (i) and (ii), curves (iii) and (iv), curves (v) and (vi) are operated at the same condition.
Photonics 11 01186 g004
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Kwon, D. Excess Intensity Noise in a Nonlinear Amplifying Loop-Mirror-Based Mode-Locked Laser from a Non-Reciprocal Phase Bias. Photonics 2024, 11, 1186. https://doi.org/10.3390/photonics11121186

AMA Style

Kwon D. Excess Intensity Noise in a Nonlinear Amplifying Loop-Mirror-Based Mode-Locked Laser from a Non-Reciprocal Phase Bias. Photonics. 2024; 11(12):1186. https://doi.org/10.3390/photonics11121186

Chicago/Turabian Style

Kwon, Dohyeon. 2024. "Excess Intensity Noise in a Nonlinear Amplifying Loop-Mirror-Based Mode-Locked Laser from a Non-Reciprocal Phase Bias" Photonics 11, no. 12: 1186. https://doi.org/10.3390/photonics11121186

APA Style

Kwon, D. (2024). Excess Intensity Noise in a Nonlinear Amplifying Loop-Mirror-Based Mode-Locked Laser from a Non-Reciprocal Phase Bias. Photonics, 11(12), 1186. https://doi.org/10.3390/photonics11121186

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop