Chinese Optics Letters, Volume. 22, Issue 10, 101402(2024)

Broadband continuously tunable Er:Yb:YAl3(BO3)4 1.5–1.6 μm laser

Yujin Chen, Yanfu Lin, Jianhua Huang, Xinghong Gong, and Yidong Huang*
Author Affiliations
  • Key Laboratory of Optoelectronic Materials Chemistry and Physics, Fujian Institute of Research on the Structure of Matter, Chinese Academy of Sciences, Fuzhou 350002, China
  • show less

    By combining the thermally-induced spectral broadening of Er3+/Yb3+ co-doped crystal and the high ratio of cavity gain to loss, a high-power broadband continuously tunable 1.5-1.6 µm laser was successfully demonstrated in an Er:Yb:YAl3(BO3)4 crystal. End-pumped by a continuous-wave 975.6 nm laser diode, three discrete tunable laser bands at 1483–1488, 1495–1503, and 1521–1612 nm were realized at an incident pump power of 7.7 W. The maximum continuously tunable bandwidth was 91 nm at 1521–1612 nm, and the maximum output power was 474 mW at 1551 nm. The output power was generally higher than 100 mW in the whole tunable range.

    Keywords

    1. Introduction

    The eye-safe 1.5–1.6 µm laser has the advantages of low transmission loss in optical fiber, high atmospheric transmission, as well as high detection sensitivity for Ge and InGaAs photodetectors[14]. Furthermore, some specific wavelengths in this spectral region have their unique application value. For example, some gases, such as H2S, CO2, CO, C2H2, and NH3, have their absorption lines at 1.5–1.6 µm[5,6]. 1560 nm laser and its frequency-doubling laser at 780 nm, which corresponds to the transition wavelength of the D2 line of the rubidium atom, can be used in atomic physics and quantum communication[7,8]. Therefore, the 1.5–1.6 µm tunable laser has important applications in many fields, such as lidar, spectroscopy, trace gas sensing, environmental monitoring, atomic physics, and broadband optical communication.

    976 nm-diode-pumped Er3+/Yb3+ co-doped material has been considered as a convenient method for obtaining a compact and low-cost 1.5–1.6 µm solid-state laser[1,9]. Up to now, using some Er3+/Yb3+ co-doped materials with broad and flat gain band at 1.5–1.6 µm as gain media, 1.5–1.6 µm tunable solid-state lasers have been investigated[1012]. A continuously tunable 1530–1560 nm laser with output power within the whole tunable range close to 40 mW has been realized in an Er:Yb:phosphate glass by using a fiber Bragg grating[10]. By using a SiO2 birefringent filter as a tuning element, a broadband tunable 1511–1593 nm laser with a maximum output power of 3 mW at 1556 nm has been demonstrated in an Er:Yb:phosphate glass[11]. Recently, a continuously broadband tunable 1503–1605 nm laser with a maximum output power of 40 mW at 1554 nm has also been obtained in an Er:Yb:Ca3NbGa3Si2O14 crystal[12]. Although the broadband tunable 1.5–1.6 µm lasers have been demonstrated in some Er3+/Yb3+ co-doped materials, the obtained maximum output power is still low, which is mainly limited by the low thermal conductivity (about 0.8Wm1K1) of the Er:Yb:glass and the low laser efficiency (<10%) of the Er:Yb:Ca3NbGa3Si2O14 crystal[1,13].

    Er:Yb:YAl3(BO3)4 (Er:Yb:YAB) crystal has high thermal conductivity of 7.7Wm1K1, high Yb3+Er3+ energy transfer efficiency of 88%, large emission cross section of 2.5×1020cm2 at 1531.5 nm, and short fluorescence lifetime of 325 µs for the I413/2 multiplet[9]. At present, the crystal has been demonstrated as an excellent 1.5–1.6 µm laser material[4,9,14]. A 1543 nm continuous-wave (CW) laser with a maximum output power of 4.55 W and a slope efficiency of 34% has been realized[14]. However, due to the ordered structure of the crystal and single-crystal field environment around Er3+ sites, the gain spectrum at 1.5–1.6 µm of the crystal contains multiple discrete narrowbands[9]. The steep and narrow gain bands limit the tunable bandwidth of the 1.5–1.6 µm laser. Therefore, the potentiality of the Er:Yb:YAB crystal as gain medium for a broadband tunable laser has been neglected up to now.

    In this work, wavelength tunability of the 1.5–1.6 µm laser is explored in the Er:Yb:YAB crystal. The temperature inside the crystal during laser operation and the high-temperature absorption spectra at 1.5–1.6 µm of the crystal are measured. Then, the influence of the crystal temperature on the gain spectrum is investigated. By combining the thermally-induced spectral broadening of the crystal and the high ratio of cavity gain to loss, a high-power broadband continuously tunable 1.5–1.6 µm laser is successfully realized in an Er:Yb:YAB crystal.

    2. Experiments

    The experimental setup for the 1.5–1.6 µm tunable laser is shown in Fig. 1. A 975.6 nm CW fiber-coupled LD with a core diameter of 105 µm was adopted. A c-cut, 1.5-mm-thick uncoated Er (1.5% in atomic fraction):Yb (12.0% in atomic fraction):YAB crystal with a cross section of 3.0mm×3.0mm was used. A pump beam with a waist diameter of about 100 µm is obtained in the crystal by using a telescopic lens system (TLS). An input mirror (IM) film with 90% transmission around 975.6 nm and 99.9% reflectivity at 1470–1620 nm was directly deposited onto the input surface of a 1.0-mm-thick sapphire crystal with a cross section of 3.0mm×3.0mm. The sapphire and Er:Yb:YAB crystals were optically contacted and mounted in a copper chamber. Three output mirrors (OMs) with different transmissions (1.2%, 2.6%, and 4.7%) at 1470–1620 nm and the same curvature radius of 100 mm were used. The flat transmission curves at 1470–1620 nm of the IM and OMs shown in Fig. 2 can effectively avoid the influence of transmission change at different wavelengths on the tunable laser performance. The cavity length was about 100 mm. A 2.0-mm-thick SiO2 birefringent filter was used as tuning element and placed in the cavity at Brewster’s angle of 60°. The laser spectrum was recorded by a spectrometer (waveScan, APE) with a resolution of 0.5 nm.

    Experimental setup for 1.5–1.6 µm Er:Yb:YAB tunable laser.

    Figure 1.Experimental setup for 1.5–1.6 µm Er:Yb:YAB tunable laser.

    Transmission curves at 1470–1620 nm of the IM and OMs.

    Figure 2.Transmission curves at 1470–1620 nm of the IM and OMs.

    3. Results and Discussion

    Based on the room-temperature absorption cross section spectrum σabs at 1450–1650 nm of a c-cut Er:Yb:YAB crystal recorded by a spectrophotometer (Lambda-950, PerkinElmer), the room-temperature emission cross section spectrum σem can be calculated by the reciprocity method[9]. Then, the room-temperature gain cross section spectrum σgain can be obtained by the formula, σgain(λ)=βσem(λ)(1β)σabs(λ)[13]. The inversion parameter β is the ratio of the Er3+ in the upper laser level to the total Er3+, and proportional to the pump power. Figure 3 shows the room-temperature gain cross section spectrum σgain of a c-cut Er:Yb:YAB crystal for different β. At β of 0.3, there is a relatively flat gain spectrum at 1570–1615 nm. With the increment of β, the gain range becomes larger, which indicates that laser oscillation can be achieved in a wider wavelength range. However, gain spectrum becomes more uneven. At β of 0.9, nine discrete and narrow gain bands are observed in the spectrum from 1475 to 1620 nm, which leads to the difficulty in realizing a continuously broadband tunable laser in the crystal.

    Room-temperature gain cross section spectra at 1450–1650 nm of a c-cut Er:Yb:YAB crystal for different β.

    Figure 3.Room-temperature gain cross section spectra at 1450–1650 nm of a c-cut Er:Yb:YAB crystal for different β.

    At a low OM transmission of 1.2% and an incident pump power of 3.0 W, the wavelength tunability of the CW laser was investigated by rotating the birefringent filter, as shown in Fig. 4(a). The intensities of different laser lines were normalized. Five discrete tunable laser bands at 1543–1544, 1550–1553, 1560–1563, 1582–1585, and 1601–1612 nm were obtained. The maximum output power of 210 mW was realized at 1603 nm. By comparing with the gain spectra for different β, the tuning curve of the output laser is consistent well with the room-temperature gain spectrum at β=0.45, as shown in Fig. 4(b). The experimental result indicates that although the gain spectrum is broadened from 1533 to 1620 nm at β=0.45, the discrete and steep narrowbands limit the bandwidth of the continuously tunable laser. The maximum continuously tunable bandwidth was only 11 nm at 1601–1612 nm. Therefore, in order to realize a continuously broadband tunable laser, some technical methods must be adopted to smooth the gain spectrum of the Er:Yb:YAB crystal.

    Wavelength tunability of the CW laser at an incident pump power of 3.0 W and an OM transmission of 1.2%. (a) Tuning laser spectra and output powers; (b) output powers and room-temperature gain spectrum at β = 0.45.

    Figure 4.Wavelength tunability of the CW laser at an incident pump power of 3.0 W and an OM transmission of 1.2%. (a) Tuning laser spectra and output powers; (b) output powers and room-temperature gain spectrum at β = 0.45.

    Due to the low fluorescence quantum efficiency (8%) of I413/2 multiplet and high quantum defect (37%) for the 976 nm-diode-pumped Er3+/Yb3+ 1.5–1.6 µm laser, a large amount of pump power is converted to heat in an Er:Yb:YAB crystal[9,14]. Therefore, the crystal will operate at high temperature during the laser oscillation. Based on the fluorescence intensity ratio (FIR) method[15], the temperature inside the crystal can be estimated when the laser is operating. Upconversion fluorescence bands originating from the H211/2I415/2 and S43/2I415/2 transitions of Er3+, which are, respectively, centered at 525 and 550 nm, can be observed in the crystal when the laser is operating. The H211/2 and S43/2 levels of Er3+ with energy separation of about 780cm1 are thermally coupled levels (TCLs)[16]. The intensity ratio of the fluorescence bands originating from these two TCLs represents the temperature at the Er3+ sites, which is called FIR thermometry[16]. On the basis of the 1.5–1.6 µm tunable laser experiment, the setup for measuring the temperature inside the crystal is depicted in Fig. 5. By using a convex lens, the upconversion fluorescence emitting from the crystal can be focused into a fiber with a core diameter of 100 µm connected to a spectrometer (HR4000, Ocean Optics). The scattering signals originating from the lasers are blocked by a filter positioned in front of the fiber.

    Experimental setup for measuring the temperature inside the Er:Yb:YAB crystal during the 1.5–1.6 µm laser operation.

    Figure 5.Experimental setup for measuring the temperature inside the Er:Yb:YAB crystal during the 1.5–1.6 µm laser operation.

    Crystal temperature T can be roughly estimated by the relationship FIR=8.67×exp(1083.6/T), in which FIR is the integral intensity ratio of fluorescence bands centered at 525 and 550 nm in the upconversion spectrum[15]. Then, the temperatures at the waist spot position of the pump beam in the crystal, where the strongest upconversion fluorescence was observed, were measured and are shown in Fig. 6 for different pump powers. At the incident pump powers of 3.0, 5.3, and 7.7 W, the crystal temperatures were about 450, 550, and 700 K, respectively. The temperature error was estimated to be ±10K by repeated measurements. The high crystal temperature will inevitably affect the spectral shape of the crystal[17].

    Temperatures at the waist spot position of the pump beam in the crystal for different pump powers when the 1.5–1.6 µm laser is operating. The insets show the upconversion fluorescence spectra at 510–570 nm recorded at some pump powers.

    Figure 6.Temperatures at the waist spot position of the pump beam in the crystal for different pump powers when the 1.5–1.6 µm laser is operating. The insets show the upconversion fluorescence spectra at 510–570 nm recorded at some pump powers.

    By using a home-made spectrometer, high-temperature absorption cross section spectra at 1450–1650 nm of a c-cut Er:Yb:YAB crystal from 300 to 800 K can be recorded. Then, based on the high-temperature emission cross section spectra calculated by the reciprocity method, the gain cross section spectra of a c-cut Er:Yb:YAB crystal for different temperatures can be obtained. As an example, the gain spectra at 1510–1630 nm of the crystal at different temperatures of 300 K, 450 K, 550 K, and 700 K for β=0.5 are shown in Fig. 7. It can be seen that with the increment of temperature, the gain spectrum of the crystal becomes flatter and smoother, although the gain cross section decreases. Therefore, the thermally-induced spectral broadening of the Er3+/Yb3+ co-doped crystal may be beneficial for realizing the continuously broadband tunable 1.5–1.6 µm laser.

    Gain cross section spectra at 1510–1630 nm of a c-cut Er:Yb:YAB crystal at different temperatures T for β = 0.5.

    Figure 7.Gain cross section spectra at 1510–1630 nm of a c-cut Er:Yb:YAB crystal at different temperatures T for β = 0.5.

    When the OM transmission was kept at 1.2%, the tunability of the CW laser at an incident pump power of 5.3 W was measured and is shown in Fig. 8(a). Figure 8(b) shows that the tuning curve of the output laser is more consistent with the gain spectrum at 550 K for β=0.65. It can be found that higher incident pump power can cause a larger inversion parameter β (i.e., higher cavity gain) and higher crystal temperature. Then, a broader and smoother gain spectrum can be obtained. Seven discrete tunable laser bands at 1483–1485, 1497, 1522–1524, 1531, 1543–1569, 1580–1592, and 1601–1612 nm were obtained. The maximum output power of 360 mW was realized at 1554 nm. The maximum continuously tunable bandwidth was 26 nm at 1543–1569 nm. Furthermore, it is worth noting that 1483–1485 nm and 1497 nm solid-state lasers are obtained for the first time, and 1483 nm is the shortest laser wavelength realized in the Er3+/Yb3+ co-doped materials via the I413/2I415/2 transition of Er3+ up to now, to the best of our knowledge. The realizing of the above laser is difficult to be explained by the room-temperature gain cross section spectrum shown in Fig. 8(b), and may originate from the thermally-induced variation of Er3+ population in the related Stark energy levels. Figure 9 shows the simplified energy level diagram of Er3+ in the YAB crystal and ion population ratios of all Stark energy levels at 300 K and 550 K[18]. At 300 K and 550 K, the ion population differences of the upper and lower energy levels for 1485 nm laser can be estimated to be proportional to 0.65×0.087(10.65)×0.239=0.0271 and 0.65×0.111(10.65)×0.183=0.0081, respectively, when β=0.65 is taken into account. Therefore, ion population inversion for 1485 nm laser can only be achieved at 550 K when β=0.65.

    Wavelength tunability of the CW laser at an incident pump power of 5.3 W and an OM transmission of 1.2%. (a) Tuning laser spectra and output powers; (b) output powers and gain spectra of the crystal at 300 and 550 K for β = 0.65.

    Figure 8.Wavelength tunability of the CW laser at an incident pump power of 5.3 W and an OM transmission of 1.2%. (a) Tuning laser spectra and output powers; (b) output powers and gain spectra of the crystal at 300 and 550 K for β = 0.65.

    Energy level diagram of Er3+ in the YAB crystal and ion population ratios of all Stark energy levels at 300 K and 550 K.

    Figure 9.Energy level diagram of Er3+ in the YAB crystal and ion population ratios of all Stark energy levels at 300 K and 550 K.

    The OM transmission was kept at 1.2% and the pump power was increased to 7.7 W; the resulting tunability of the CW laser is shown in Fig. 10(a). The tuning curve is more consistent with the gain spectrum at 700 K for β=0.75 shown in Fig. 10(b). Based on the thermally-induced spectral broadening of the crystal and the high ratio of cavity gain to loss, a broadband continuously tunable 1.5–1.6 µm laser is demonstrated in the crystal. Three discrete tunable laser bands at 1483–1488, 1495–1503, and 1521–1612 nm were obtained. The total tunable bandwidth was 104 nm, which covers the whole C (1530–1565 nm) band as well as parts of the S (1460–1530 nm) and L (1565–1625 nm) bands in the optical communication window. The maximum continuously tunable bandwidth was 91 nm at 1521–1612 nm, and the maximum output power was 474 mW at 1551 nm. The output power was generally higher than 100 mW in the whole tunable range. The absence of laser oscillation at 1503–1521 nm may be caused by the significant difference of gain cross sections at 1503 and 1521 nm. Then, the adoption of a thicker SiO2 wafer with a higher wavelength selectivity may realize the laser oscillation between 1503 and 1521 nm in the future. Compared with those (82–102 nm bandwidth and about 40 mW output power) of the Er3+/Yb3+ co-doped glass and Ca3NbGa3Si2O14 crystal[1012], the similar tunable bandwidth but 1 order of magnitude higher output power was realized in the Er:Yb:YAB crystal. Consequently, a high-power broadband continuously tunable 1.5–1.6 µm laser has been realized in an Er:Yb:YAB crystal for the first time. As an example, Fig. 10(c) shows the spectrum as well as 2D and 3D images of the transversal profile for the 1551 nm laser beam recorded by a Pyrocam III camera (Ophir Optronic Ltd.). The full width at half-maximum (FWHM) of the laser line was about 0.7 nm, which is close to the resolution (0.5 nm) of the spectrometer. The quality factor M2 of output beam was fitted to be 1.7.

    Wavelength tunability of the CW laser at an incident pump power of 7.7 W and an OM transmission of 1.2%. (a) Tuning laser spectra and output powers; (b) output powers and gain spectra of the crystal at 300 K and 700 K for β = 0.75; (c) spectrum as well as images of the transversal profile for the laser beam.

    Figure 10.Wavelength tunability of the CW laser at an incident pump power of 7.7 W and an OM transmission of 1.2%. (a) Tuning laser spectra and output powers; (b) output powers and gain spectra of the crystal at 300 K and 700 K for β = 0.75; (c) spectrum as well as images of the transversal profile for the laser beam.

    When the OM transmission was increased to 2.6% and 4.7%, the tuning spectra of the CW lasers at an incident pump power of 7.7 W were measured and are shown in Figs. 11(a) and 11(b), respectively. With the increment of the cavity loss caused by the higher OM transmission, which leads to a reduction of the ratio of cavity gain to loss, the continuously tunable bandwidth of the 1.5–1.6 µm laser decreases. For OM transmissions of 2.6% and 4.7%, the maximum continuously tunable bandwidths were 65 nm at 1543–1608 nm and 21 nm at 1543–1564 nm, respectively. The corresponding maximum output powers were 670 mW at 1551 nm and 664 mW at 1550 nm, respectively. Furthermore, at the OM transmission of 2.6%, the output powers at the short laser wavelengths of 1485 and 1499 nm reached up to 270 and 325 mW, respectively. Using a Glan–Taylor polarizer, the output lasers at all wavelengths were measured to be linearly polarized for all the OM transmissions because of the adoption of a birefringent filter placed at Brewster’s angle in the cavity.

    Wavelength tunability of the CW laser at an incident pump power of 7.7 W for different OM transmissions T. (a) T = 2.6%. (b) T = 4.7%.

    Figure 11.Wavelength tunability of the CW laser at an incident pump power of 7.7 W for different OM transmissions T. (a) T = 2.6%. (b) T = 4.7%.

    4. Conclusion

    By combining the thermally-induced spectral broadening of the Er:Yb:YAB crystal and the high ratio of cavity gain to loss, a high-power broadband continuously tunable 1.5–1.6 µm laser is successfully realized in the Er3+/Yb3+ co-doped material for the first time. Compared with those of the commercial Er:Yb:phosphate glass, the larger continuously tunable bandwidth and 1 order of magnitude higher output power can be obtained in the Er:Yb:YAB crystal.

    [1] B. Denker, B. Galagan, S. Sverchkov et al. Erbium (Er) glass lasers. Handbook of Solid-State Lasers, 341(2013).

    Tools

    Get Citation

    Copy Citation Text

    Yujin Chen, Yanfu Lin, Jianhua Huang, Xinghong Gong, Yidong Huang, "Broadband continuously tunable Er:Yb:YAl3(BO3)4 1.5–1.6 μm laser," Chin. Opt. Lett. 22, 101402 (2024)

    Download Citation

    EndNote(RIS)BibTexPlain Text
    Save article for my favorites
    Paper Information

    Category: Lasers, Optical Amplifiers, and Laser Optics

    Received: Mar. 13, 2024

    Accepted: May. 15, 2024

    Published Online: Oct. 12, 2024

    The Author Email: Yidong Huang (huyd@fjirsm.ac.cn)

    DOI:10.3788/COL202422.101402

    CSTR:32184.14.COL202422.101402

    Topics