See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/330910962

Passively Q-switched fiber lasers based on pure water as the saturable

absorber

Article  in  Optics Letters · February 2019

DOI: 10.1364/OL.44.000863

CITATIONS

11READS

826

5 authors, including:

Some of the authors of this publication are also working on these related projects:

Single-longitudinal-mode fiber lasers View project

Ultrafast fiber lasers View project

Tianhao Xian

Shanghai Jiao Tong University

16 PUBLICATIONS   37 CITATIONS   

SEE PROFILE

Li Zhan

Shanghai Jiao Tong University

186 PUBLICATIONS   2,673 CITATIONS   

SEE PROFILE

Lirun Gao

Shanghai Jiao Tong University

16 PUBLICATIONS   41 CITATIONS   

SEE PROFILE

Wenyan Zhang

Shanghai Jiao Tong University

15 PUBLICATIONS   37 CITATIONS   

SEE PROFILE

All content following this page was uploaded by Li Zhan on 19 April 2021.

The user has requested enhancement of the downloaded file.

Passively Q-switched fiber lasers based on purewater as the saturable absorberTIANHAO XIAN, LI ZHAN,* LIRUN GAO, WENYAN ZHANG, AND WENCHAO ZHANG

State Key Laboratory of Advanced Optical Communication Systems and Networks, School of Physics and Astronomy,Shanghai Jiao Tong University, Shanghai 200240, China*Corresponding author: lizhan@sjtu.edu.cn

Received 30 November 2018; revised 9 January 2019; accepted 14 January 2019; posted 15 January 2019 (Doc. ID 353331);published 6 February 2019

We propose and demonstrate a passively Q-switchedEr-doped fiber laser based on pure water as the saturableabsorber (SA). The SA is made of two optical ferrulesmatched with a cannula, and the gap between the end-facetsis filled with pure water. The nonlinear response of this SAhas been characterized, and stable Q-switching operation at1558.03 nmhas been achieved. Themaximum output poweris 21.1 mW with 65.0 kHz repetition rate. The duration is1.44 μs, and the pulse energy reaches 324.8 nJ. To the best ofour knowledge, this is the first demonstration of the pas-sivelyQ-switched laser with pure water as the SA. It providesfurther evidence of the possibility of liquid as an effectiveSA for pulsed lasers. © 2019 Optical Society of America

https://doi.org/10.1364/OL.44.000863

Owing to their versatile applications to remote sensing, materialprocessing, photoacoustic imaging, medicine, and laser process-ing, Q-switched fiber lasers have attracted much interest in re-cent years [1–4]. There are two primary approaches to realizeQ-switched lasers, namely, active Q-switching [5] and passiveQ-switching [6]. Compared to the active Q-switched scheme,passive Q-switched fiber lasers possess advantages of compact-ness, simplicity, and flexibility in design. Passive Q-switchingoperation of lasers is usually produced by employing SAs.The response time of the SAs for Q-switching need not beshorter than the round-trip time, but should be shorter thanthe lifetime of the upper level of a gain medium, because thepulse duration is determined by the depletion time after satu-ration [7]. Usually,Q-switching demands themodulation depthof the SA to be large enough to change the cavity loss, and de-mands that the SA can support high-pulse-energy operation.Therefore, an ideal SA for Q-switching to maintain stable pulseoperation should possess the features of large modulation depth,high damage threshold, and good thermal stability.

Passively Q-switched Er-doped fiber lasers (EDFLs) havebeen well investigated using different SAs such as a semicon-ductor saturable absorber mirror (SESAM) [8,9], graphene[10,11], carbon nanotubes (CNTs) [12,13], gold nanocrystal[14], rare-earth-doped fibers [15], and even alcohol [16,17].

However, SESAMs have narrow wavelength tuning ranges(10s of nanometers), and are complex components for fabrica-tion [18]. Graphene and CNT SAs perform with low saturationpower, low cost, and broadband operation, but they possesssuch a low damage threshold that they are easily damaged inQ-switching operation [19]. Rare-earth-doped fibers havethe drawback that the pulse characteristics are difficult to vary,as they depend on the fiber fabrication process [20]. Recently,ethanol has been introduced as a SA for Q-switching orfor mode locking [16,21,22], but ethanol evaporates easilyand is organic material, which contains other bonds besidesthe oxygen–hydrogen bond (O–H).

As a matter basic to biology, water is ubiquitous in theworld. It is a small molecule that contains only three atoms.The chemical bond in water molecule is only the O–H bond,which has a wide absorption band in 1550 nm. As a SA, purewater possesses the advantages of large liquidity, good thermaldiffusivity, high damage threshold, and a self-healing feature.Therefore, water SAs are possibly a valid way to realizeQ-switching or mode-locking operation in lasers.

In this Letter, we propose and demonstrate pure water as aSA for Q-switched EDFLs. We believe that it is the first timesuch a SA has been used in Q-switched lasers. The maximumoutput power is 21.1 mW, while the repetition rate is 65.0 kHz.The pulse duration is 1.44 μs, and the pulse energy is 324.8 nJ.By optimizing the layer thickness, the modulation depth ofwater SA can be flexibly changed for different requirements. Theadvantage of water SA is that it is easy to obtain, hard to beevaporated, and has a high damage threshold, which may offerimportant contributions to develop short-pulse fiber lasers.

Figure 1(a) depicts the measured absorption spectrum ofwater (for a more accurate absorption spectrum, refer to[23,24]), indicating that it has a broad absorption band around1550 nm and that it can be utilized as a SA in EDFL. Wateris a micro-molecule with only two hydrogen atoms and oneoxygen atom and has only the O–H bond, which excludesthe influence of other bonds such as carbon–oxygen (C–O)and carbon–hydrogen (C–H) bonds and guarantees theabsorption coming from the interaction of a laser with theO–H. Moreover, the O–H bond is difficult to be separated,which is another advantage of water SA for Q-switching.

Letter Vol. 44, No. 4 / 15 February 2019 / Optics Letters 863

0146-9592/19/040863-04 Journal © 2019 Optical Society of America

The absorption band covering 1550 nm corresponds to thestretching vibration of the O–H bond, and the band near1890 nm is consistent with the flexural vibration of the O–Hbond [24,25], which provides adequate absorption for 1550 nmand 1890 nm bands to Q-switched and/or mode-locked EDFLor Tm-doped fiber lasers [26]. In addition, water has a wideabsorption band around 3 μm, and owing to the fast responsetime of this absorption band, it can contribute to pulsed fiberlasers in the 3 μm waveband [25,27].

In our experiment, the SA was made of two FC/PC opticalferrules matched by the cannula, as shown in the inset in Fig. 2.The ceramic cannula was taken from a FC/PC fiber connectoras a matching structure. The end-facets of the optical ferruleswere wiped with pure water before being immersed into purewater in a dish. Then they were inserted into the cannula, andthe gap between the two ferrules was filled with water. This stepwas operated in pure water to prevent bubbles in the gapbetween connectors. The water was deionized water with purityof 99.99%. Water in the gap acted as the SA because of itsabsorption near the 1550 nm wavelength region. The thicknessof the water layer can be easily changed by varying the gap, so itis easy to obtain various SAs with different modulation depths,which is a unique advantage of liquid SAs.

To characterize the nonlinear response of the water SA, atypical balanced two-detector measurement system was em-ployed [28,29]. The probe light was a homemade mode-lockedfiber laser source with pulse duration of 1.5 ps at 1566 nm andrepetition rate of 44.6 MHz. 10% of the output power wasmonitoring the input power, while the 90% residual signalwas incident into the water SA. The measured transmission rateof the SA is shown in Fig. 1(b). Typical saturable absorptionwas observed, and the experimental data are well fitted with thesaturable absorption formula of the transmission rate α as

α � α01� I∕I sat

� αns, (1)

where α0 is the modulation depth, αns is the non-saturableabsorbance, I is the input intensity, and I sat is the saturatingintensity. The modulation depth of the SA is 5%, which isadequate for Q-switching or mode locking [30]. The thicknessof the gap is measured using a microscope, as shown in the insetin Fig. 2. Compared with the 0.5 mm aperture of the cannula,the thickness of the gap is ∼112 μm. The relaxation time of thewater is on the order of picoseconds, based on the transitionbetween the vibrational energy levels of the water molecules[31]. The response time of the SA is much shorter than the life-time of the upper level in the gain medium, and thus this waterSA provides an adequate property forQ-switching in fiber lasers.

To evaluate the performance of the water SA, a Q-switchedEDFL has been constructed as shown in Fig. 2. A piece of EDFwith length of 3 m and absorption rate of 21 dB/m at 1530 nmwas utilized as the gain medium, and it was forward pumped bya 980 nm laser diode (LD) through a 980/1550 wavelengthdivision multiplexer (WDM). A 20:80 optical coupler (OC)extracted 20% of the generated laser for measurements. Apolarization independent isolator (ISO) was placed betweenthe SA and the OC to ensure unidirectional lasing in the cavity.A polarization controller (PC) was inserted to optimize the laseroutput characterizing the polarization statuses of theQ-switched pulses. The total cavity length was 12.7 m. Theoutput spectrum was measured with an optical spectrum ana-lyzer (Yokogawa AQ6370), while the pulse train from the laserwas monitored with a 40 GHz photodetector and visualized onan oscilloscope (RTO 1002, 2 GHz).

When the laser cavity contained no SA, we increased thepump power from zero to the maximum power together withchanging polarization statuses in a full range by adjusting thePC; only continuous-wave (CW) operation was observed. Thisexcluded self-Q-switching of the EDFL [32]. Nevertheless,after inserting the SA into the laser cavity, the remarkable pas-sive Q-switching operation occurred at the pump power of200 mW. Figure 3 shows the typical oscilloscope traces of

Fig. 1. (a) Absorption spectrum of pure water. (b) Nonlinearresponse of the pure water SA at 1550 nm wavelength.

Fig. 2. Experimental set up of Q-switched fiber laser using purewater as SA. The inset shows the schematic image of the SA.

Fig. 3. Trains of the Q-switched EDFL under different pumppowers.

864 Vol. 44, No. 4 / 15 February 2019 / Optics Letters Letter

the Q-switched pulse trains under different pump powers.In the process of varying the pump power, the pulse outputremains stable, and the repetition rates are in dozens of kilo-hertz range. The pulse trains are shown for pump powers inthe range from 230 mW to 370 mW, where stable passivelyQ-switched pulses are observed. The CW operation was ob-served for the pump power below 150 mW. The thresholdfor the Q-switching is high, owning to the insertion loss ofthe SA and the absorption in the water layer. The Q-switchingoperation can remain stable for hours for the whole test process,and there is no need to refill water into the gap between eachoperation because the SA is immersed in pure water. Duringthe operation for hours, we observed no bubble in the gap.

The waveform and the optical spectrum of the Q-switchedpulses at pump power of 290 mW are shown in Fig. 4. Thepulse repetition rate is 62.2 kHz, corresponding to the timeinterval of 16.08 μs, as shown in Fig. 3(b). Figure 4(a) showsthe single pulse shape of the Q-switched pulses and exhibitsthat the pulse duration is 1.61 μs. The central wavelengthof the laser signal located at 1558.03 nm with the 3 dB band-width of 1.24 nm is shown in Fig. 4(b). The pulse duration andthe spectrum width present a typical feature of Q-switchedfiber lasers. Varying polarization status in the cavity, the pulsetrain and spectrum display almost no change, which manifeststhat the Q-switching operation of this laser is polarizationinsensitive.

The average output power of the laser pulses and the single-pulse energy versus the pump power are sketched in Fig. 5(a),while Fig. 5(b) presents the pulse duration and pulse repetitionrate as functions of pump power. The output power varies from13.0 mW to 21.1 mW by increasing the pump power from230 mW to 370 mW, while the repetition rate changesfrom 55.7 kHz to 65.0 kHz. Unlike mode-locked operation,the repetition rate of Q-switched pulses depends on the pump

power rather than the cavity length, since pulse generation de-pends on the saturation of the SA. When increasing the pumppower, more gain is provided to saturate the SA, and the SA issaturated quicker, which results in the increase in the repetitionrate. The single-pulse energy scales from 234.0 nJ up to324.8 nJ, and the pulse duration decreases gradually from1.90 μs to 1.44 μs. The maximum output power is 21.1 mWfor 65.0 kHz repetition rate at the pump power of 370 mW,corresponding to maximum pulse energy of 324.8 nJ.

The water SA performs the intrinsic advantages of the liquidwith an excellent self-healing feature, good thermal diffusivity,and high damage threshold. The chemical bond in water hasonly the O–H bond, removing the disturbance of other bondssuch as C–O and C–H in alcohol. In addition, the thickness ofthe water layer can be easily controlled by changing the gap,which facilitates obtaining SAs with various modulationdepths for different applications. Hence, by carefully changingthe modulation depth, the mode-locking operation can beobtained in EDFL. Also, following water’s absorption bandsaround 1890 nm and 3 μm, together with the picosecond’s life-time of the vibration level of O–H, water is an alternative SA inTm-doped fiber lasers or 3 μm fiber lasers for mode locking orQ-switching.

As an excellent solvent, water can dissolve a quantity ofinorganic matter, which provides the possibility to fabricatevarious aqueous solution SAs. The aqueous solution SA canoperate in the wavelength range depending on the solute, mak-ing it possible to employ this category of SA in other kinds oflasers, if the working wavelength is in the absorption band ofthe aqueous solution. For instance, by adjusting the gap width,solution SA can be employed in single-frequency fiber lasers[33]. Moreover, one can fabricate all-fiber-based SAs withwater-filled or solution-filled photonic crystal fiber to obtainshort-pulse operation in all-fiber lasers [34]. Certainly, userscan fabricate transparent cells filled with solution or water asSAs and insert them into solid-state lasers for mode lockingor Q-switching [35,36].

In conclusion, a water-based passivelyQ-switched EDFL hasbeen proposed and experimentally demonstrated. StableQ-switched pulses located at 1558.03 nm have been achieved.This Q-switched EDFL can deliver a maximum average outputpower of 21.1 mW at 370 mW pump power with the pulse rep-etition rate of 65.0 kHz. The single-pulse energy is 324.8 nJ, andpulse duration is 1.44 μs. The vibration absorption of the O–Hbond in water molecules contributes to the Q-switching oper-ation. To the best of our knowledge, this is the first demonstra-tion of a passively Q-switched laser using pure water as the SA.Considering the picoseconds response time of water vibrationlevels and variable modulation depths, mode-locked fiber lasersbased on water SA operating in 1.55 μm range, 2 μm range, andeven 3 μm range can be imagined.

Funding. National Natural Science Foundation of China(NSFC) (11874040); Foundation for Leading Talents ofMinhang, Shanghai.

REFERENCES

1. J. Y. Huang, W. C. Huang, W. Z. Zhuang, K. W. Su, Y. F. Chen, andK. F. Huang, Opt. Lett. 34, 2360 (2009).

Fig. 4. (a) Profile of Q-switched pulse at pump power of 290 mW.(b) Corresponding Q-switched optical spectrum.

Fig. 5. (a) Average output power and single-pulse energy as a func-tion of pump power. (b) Pulse duration and pulse repetition rate versuspump power.

Letter Vol. 44, No. 4 / 15 February 2019 / Optics Letters 865

2. M. Skorczakowski, J. Swiderski, W. Pichola, P. Nyga, A. Zajac, M.Maciejewska, L. Galecki, J. Kasprzak, S. Gross, A. Heinrich, andT. Bragagna, Laser Phys. Lett. 7, 498 (2010).

3. Z. Luo, Y. Huang, M. Zhong, Y. Li, J. Wu, B. Xu, H. Xu, Z. Cai, J. Peng,and J. Weng, J. Lightwave Technol. 32, 4679 (2014).

4. Z. Piao, L. Zeng, Z. Chen, and C. S. Kim, Appl. Phys. Lett. 108,143701 (2016).

5. Q. Zhao, Z. Wu, Z. Zhang, W. Lin, C. Li, X. Guan, T. Tan, C. Yang, H.Cheng, J. Gan, Z. Feng, M. Peng, Z. Yang, and S. Xu, Opt. Express26, 17000 (2018).

6. C. E. Preda, G. Ravet, and P. Mégret, Opt. Lett. 37, 629 (2012).7. U. Keller, K. J. Weingarten, F. Kärtner, D. Kopf, B. Braun, I. D. Jung,

R. Fluck, C. Hönninger, N. Matuschek, and J. Aus der Au, IEEE J. Sel.Top. Quantum Electron. 2, 435 (1996).

8. R. Paschotta, R. Häring, E. Gini, H. Melchior, U. Keller, H. L.Offerhaus, and D. J. Richardson, Opt. Lett. 24, 388 (1999).

9. B. Braun, F. X. Kärtner, U. Keller, J. P. Meyn, and G. Huber, Opt. Lett.21, 405 (1996).

10. Z. Luo, M. Zhou, J. Weng, G. Huang, H. Xu, C. Ye, and Z. Cai, Opt.Lett. 35, 3709 (2010).

11. Y. Tang, X. Yu, X. Li, Z. Yan, and Q. J. Wang, Opt. Lett. 39, 614(2014).

12. S. Kivistö, T. Hakulinen, A. Kaskela, B. Aitchison, D. Brown, A.Nasibulin, E. Kauppinen, A. Härkönen, and O. Okhotnikov, Opt.Express 17, 2358 (2009).

13. A. Martinez and Z. Sun, Nat. Photonics 7, 842 (2013).14. T. Jiang, Y. Xu, Q. Tian, L. Liu, Z. Kang, R. Yang, G. Qin, and W. Qin,

Appl. Phys. Lett. 101, 151122 (2012).15. V. V. Dvoyrin, V. M. Mashinsky, and E. M. Dianov, Opt. Lett. 32, 451

(2007).16. Z. Wang, L. Zhan, M. Qin, J. Wu, L. Zhang, Z. Zou, and K. Qian,

J. Lightwave Technol. 33, 4857 (2015).17. B. Ibarra-Escamilla, M. Durán-Sánchez, B. Posada-Ramírez, R. I.

Álvarez-Tamayo, J. Alaniz-Baylón, M. Bello-Jiménez, P. Prieto-Cortés, and E. A. Kuzin, IEEE Photon. Technol. Lett. 30, 1768 (2018).

18. U. Keller, Nature 424, 831 (2003).19. A. Martinez, K. Fuse, B. Xu, and S. Yamashita, Opt. Express 18,

23054 (2010).20. T. Y. Tsai, Y. C. Fang, and S. H. Hung, Opt. Express 18, 10049

(2010).21. Z. Wang, L. Zhan, J. Wu, Z. Zou, L. Zhang, K. Qian, L. He, and X.

Fang, Opt. Lett. 40, 3699 (2015).22. Y. Hsu, C. Yeh, and C. Chow, Opt. Quantum Electron. 49, 360

(2017).23. J. A. Curcio and C. C. Petty, J. Opt. Soc. Am. 41, 302 (1951).24. J. E. Bertie and Z. Lan, Appl. Spectrosc. 50, 1047 (1996).25. S. Burikov, T. Dolenko, S. Patsaeva, Y. Starokurov, and V. Yuzhakov,

Mol. Phys. 108, 2427 (2010).26. C. Gao, Z. Wang, H. Luo, and L. Zhan, J. Lightwave Technol. 35, 2988

(2017).27. C. Wei, X. Zhu, F. Wang, Y. Xu, K. Balakrishnan, F. Song, R. A.

Norwood, and N. Peyghambarian, Opt. Lett. 38, 3233 (2013).28. S. Chen, Y. Chen, M. Wu, Y. Li, C. Zhao, and S. Wen, IEEE Photon.

Technol. Lett. 26, 987 (2014).29. X. Wang, Z. Luo, H. Liu, M. Liu, A. Luo, and W. Xu, Appl. Phys. Lett.

105, 161107 (2014).30. Z. Sun, T. Hasan, F. Torrisi, D. Popa, G. Privitera, F. Wang, F.

Bonaccorso, D. M. Basko, and A. C. Ferrari, ACS Nano 4, 803(2010).

31. S. Woutersen, U. Emmerichs, H. Nienhuys, and H. Bakker, Phys.Rev. Lett. 81, 1106 (1998).

32. T. Tsai and Y. Fang, Opt. Express 17, 21628 (2009).33. R. Poozesh, K. Madanipour, and P. Parvin, J. Lightwave Technol. 36,

4880 (2018).34. Z. Liu, X. He, and D. Wang, Opt. Lett. 36, 3024 (2011).35. T. L. Huang, S. C. Li, C. H. Tsou, H. C. Liang, K. F. Huang, and Y. F.

Chen, Opt. Lett. 43, 5753 (2018).36. C. Luan, K. Yang, J. Zhao, S. Zhao, T. Li, H. Zhang, J. He, L.

Song, T. Dekorsy, M. Guina, and L. Zheng, Opt. Lett. 42, 839(2017).

866 Vol. 44, No. 4 / 15 February 2019 / Optics Letters Letter

View publication statsView publication stats