Open Access
Add to BibSonomyAbstract
A continuous-wave microchip dual-frequency laser (DFL) with well balanced intensity was presented. In order to obtain such a balanced intensity distribution of the two frequency components, the DFL wavelengths were precisely tuned and spectrally matched with the emission cross section (ECS) spectrum of the gain medium by employing a temperature controller. Finally, when the heat sink temperature was controlled at −5.6°C, a 264 mW DFL signal was achieved with frequency separation at 67.52 GHz and intensity balance ratio (IBR) at 0.991.
© 2016 Optical Society of America
1. Introduction
Over the past decades, the photonics-based coherent radio-frequency (RF) signal sources are widely developed and show great potentials in the applications of wireless communication, material spectroscopy, frequency reference and hybrid lidar-radar fields [1–7]. Among various methods of producing photonic RF signal with high spectral purity, the frequency beating at photodiode based on dual-frequency lasers (DFLs) with RF frequency separations currently provides an effective way [8–11].
Nd3+ doped laser materials, which are appreciated for their large emission cross sections (ECSs) and large absorption cross sections (ACSs), nowadays become great gain media for the DFLs with large frequency separations [12–19]. For instance, the dual-polarization-mode microchip DFLs with frequency separations up to 150 GHz are obtained by employing Nd:YAG materials [14–16]. And the dual-longitudinal-mode microchip DFLs with frequency separations up to sub-THz are realized with Nd:YVO4 crystals [17–19].
With the obtained DFL signals for heterodyne frequency beating, the relative intensities of the two frequency components are very important factors for the beat-noting efficiency. Some efforts have been made to improve the intensity balance ratio (IBR) [20], but a few Nd3+ based DFLs are still with unbalanced intensity output [21–24]. Since the relative intensities of the Nd3+ based DFLs are strongly due to the wavelength dependent ECS values of the two frequency components, the IBR can be tuned by shifting the wavelengths in the unflat ECS spectral profiles (e.g. lorentz, gauss or dual-band functions [24,25]). In some past researches, considering the inherent thermal characteristics, the wavelengths shifting and the IBR tuning of the microchip DFLs are easily realized by controlling the laser cavity temperature [19, 25]. However, the temperature dependent ECS spectral evolutions of the gain media are often excluded from these researches, which also affect the two ECS values [26,27]. Above all, the ECS values are not only dependent on the laser wavelengths but also on the ECS spectra of gain media. For the aim of balanced intensity distribution of the microchip DFL, it is necessary to investigate the relationship among the temperature dependent laser wavelengths, the ECS spectra of gain media, and their interaction to the relative intensities.
In section 2, the microchip DFL output spectra at different temperature are studied, including the laser wavelengths and the relative intensities. In section 3, the temperature dependent ECS spectra of an uncoated gain medium are analyzed. In section 4, after comparing the DFL spectra and ECS spectra, the principle of the relative intensities variation is investigated. The conclusion is drawn in section 5.
2. The temperature dependent microchip DFL output spectra
The experimental setup used for the microchip DFL output spectra acquisition at different temperature conditions were described in Fig. 1.
An 808 nm fiber-coupled LD was used as a pump source, and the maximum pump power was 5.0 W. The fiber core diameter was 400 μm and the numerical aperture (NA) was 0.22. The pump beam was focused on the gain medium through a GRIN Lens (Ø1.8 mm, 0.29 pitch, uncoated) with 85% coupling efficiency. The output DFL beam was split into two paths by a partial mirror, and then separately fed into a power meter (30A-BB-18, Ophir Inc.) and a high resolution (0.01 nm) optical spectrum analyzer (OSA, AQ 6319, Yokogawa Inc.). The gain medium was an A-cut, 1.0 at.% Nd:YVO4 crystal with a dimension of 3.0 mm × 3.0 mm × 1.0 mm. The polarization state of the lasing light is linear polarization state. Both the input and output coupling mirror were directly coated on the ends of the Nd:YVO4 crystal, forming a monolithic laser resonator. The input mirror had high reflection (HR) at 1064 nm and antireflection (AR) at 808 nm. The output mirror had partial reflectivity of 90% at 1064 nm. To enhance thermal contact conductance and quickly remove the heat deposition, the crystal was wrapped by indium foil and mounted in a copper heat sink.
With the pump power at 1.0 W, the heat sink temperature (Tc) was precisely controlled by a thermoelectric cooler (TEC) from −4°C to 96°C. The output spectra of the microchip DFL with 1.0 mm cavity length were shown in Fig. 2(a). The spectra was normalized by the maximal laser intensity at Tc = −4°C. Different from the orthogonal polarization two-frequency Nd3+:YAG microchip laser [28], the polarization state of the Nd:YVO4 DFL remained linear polarization and no intra-mode was observed. When Tc was lower than 36°C, the modes λ0 and λ1 both oscillated in the cavity. While Tc reached 36°C, another λ-1 started to oscillate, and the tri-frequency laser oscillation arose. When Tc exceeded 36°C, the mode λ1 shifted out of the gain bandwidth, and only the modes λ0 and λ-1 oscillated in the laser cavity. The phenomenon of mode λ1 shifting out and mode λ-1 shifting in the gain bandwidth was also called “mode hopping” somewhere. Figure 2(b) showed the microchip DFL wavelengths. The λ1 shifted from 1064.53 nm to 1064.84 nm with Tc from −4°C to 36°C, the λ-1 shifted from 1064.34 nm to 1064.73 nm with Tc from 36°C to 96°C, and the λ0 shifted from 1064.28 nm to1065.01 nm with Tc from −4°C to 96°C. Over the whole temperature range investigated, the wavelengths, λ-1, λ0 and λ1, all presented linear red-shifting with Tc increasing, and the red-shifting rate were measured as 6.75 pm/°C, 7.40 pm/°C and 7.96 pm/°C, respectively.
Fig. 2 (a) Normalized laser spectra with Tc from −4°C to 96°C; (b) Laser wavelengths λq and SEC wavelength λSEC vs. Tc, q = −1, 0, 1. SEC: spectral envelope centre.
Also, the thermally induced wavelengths red-shifting rate of DFL also can be revealed by the calculation with thermal expansion coefficient αe and temperature dependent refractive index dn/dT of the gain medium [19]:
In the Eq. (1), depending on the specifications of A-cut Nd:YVO4 crystal, the wavelength red-shifting rate was calculated as 6.15 pm/°C.
By investigating of the DFL relative intensities in Fig. 2(a), it clearly showed a decreasing trend of the overall laser intensity with Tc increasing. What was more, when Tc was lower than 36°C, the intensity of the short wavelength mode λ0 was larger than the long wavelength mode λ1. When Tc was between 36°C~71°C, the mode λ1 disappeared and the mode λ0 became the long wavelength mode. The intensity of the mode λ0 was keeping larger than the new emerged short wavelength mode λ-1. When Tc exceeded 71°C, the intensity of the long wavelength mode λ0 decreased to the value smaller than the short wavelength mode λ-1. When Tc was controlled at −4°C and 71°C, the IBR of P(λ1)/P(λ0)|Tc = −4°C and P(λ0)/P(λ-1)|Tc = 71°C were calculated as 0.947 and 0.972, respectively. When Tc was controlled at 36°C, the IBR of P(λ1)/P(λ-1)|Tc = 36°C was calculated as 0.912. By connecting the average wavelength of‾λ = λ0 + λ1, λ-1 + λ1, λ-1 + λ0 at the temperature points of −4°C, 36°C and 71°C, the laser spectral envelope centre (SEC) line can be approximately defined, showed in Fig. 2(b). The relationship between λSEC and Tc was linearly fit as: λSEC = 3.88 × 10−3Tc + 1064.43 (nm). Here, λSEC was the centre wavelength of the DFL spectral envelope.
3. The temperature dependent Nd:YVO4 crystal ECS spectra
In order to study the temperature dependent ECS spectral evolution of the Nd:YVO4 crystal, the fluorescence spectra with different temperature were measured. The experimental setup used for fluorescence spectra acquisition was very similar to the laser spectra acquisition experiment. Only the monolithic microchip DFL was replaced by an uncoated Nd:YVO4 crystal with same dimension and Nd3+ doping concentration. To guarantee a minimal pump induced temperature increasing of the uncoated Nd:YVO4 crystal, a lower pump power of 100 mW was adopted in the experiment. With the indium foil wrapped, the uncoated Nd:YVO4 crystal temperature TN was approximately equal to the heat sink temperature Tc, expressed as TN = Tc. When Tc was controlled from 0°C to 100°C, the π-polarization fluorescence spectra were also collected by the OSA.
The fluorescence spectrum with Tc at 50°C was shown in Fig. 3(a) over the spectral range from 1054.30 nm to 1094.30 nm. The fluorescence spectrum had 3 spectral peaks, and the main peak wavelength was around 1064 nm. Within Tc from 0°C to 100°C, the temperature dependent ECS spectra was deduced from the Nd:YVO4 crystal fluorescence spectra by the modified Fuchtbauer–Ladenburg (FL) equation [26]:
Fig. 3 (a) Normalized fluorescence spectra vs. wavelength at Tc = 50°C; (b) Normalized ECS spectra vs. wavelength at Tc = 0°C, 50°C, and 100°C; (c) The ECS spectral peak wavelength λσmax with Tc from 0 to 100°C; (d) Normalized ECS spectral peak value σeff and FWHM with Tc from 0 to 100°C.
Here, the σ(λ,T) is the temperature dependent ECS spectrum, λ* is the average emission wavelength, c is the speed of light in vacuum, n(T) is the temperature dependent optical index of, τrad(T) is the temperature dependent radiative lifetime and I(λ,T) is the temperature dependent fluorescence intensity. Neglecting the refractive index and radiation lifetime change, the ECS spectral evolutions only depend on the temperature dependent fluorescence spectra.
By Eq. (2), the ECS spectra the uncoated Nd:YVO4 crystal around 1064 nm for Tc at 0°C, 50°C, and 100°C were curved in Fig. 3(b). The spectra were normalized by the maximal ECS value at 0°C. The experimental results showed that when increased Tc the ECS spectral peak wavelength λσmax red-shifted, the ECS spectral peak value σeff decreased, and the ECS spectral full width at half maximum (FWHM) broadened. As showed in Fig. 3(c), the λσmax red-shifted from 1064.22 nm to 1064.61 nm when Tc increased from 0°C to 100°C. The relationship was fitted as the expression λσmax = 3.84 × 10−3Tc + 1064.22 (nm) and the λσmax red-shifting rate was 3.84 pm/°C. In Fig. 3(d), the normalized ECS spectral peak value σeff(Tc) decreased 31.9% with Tc increased from 0°C to 100°C. The σeff(Tc) = max[σ(Tc)]/max[σ(0°C)], and the max[σ(0°C)] corresponded to the ECS spectral peak value at 0°C. The ECS spectral FWHM vs. Tc was also shown in Fig. 3(d). The FWHM linearly broadened with Tc increasing from 0°C to 100°C, and the broadening rate was fitted as 7.86 pm/°C. The overall 100°C temperature increasing about resulted in 0.79 nm broadening of the ECS spectral FWHM.
4. The principle of the relative intensities variation of the DFL signals
For analyzing the principle of the relative intensities variation of the DFL signals, the relationship among the DFL wavelengths, ECS spectra and their interaction to the output relative intensities were comprehensively discussed. According to the fitted expressions above, the achieved wavelength λSEC = 3.88 × 10−3Tc + 1064.43 (nm) and λσmax = 3.84 × 10−3Tc + 1064.22 (nm) were curved in Fig. 4(a). With Tc from −4°C to 100°C, the red-shifting rate of λSEC and λσmax were both 3.8x pm/°C. Here, for simple calculation, the both red-shifting rate were assumed to be the average value 3.86 pm/°C, approximately. Therefore, the wavelength difference Δλ = λSEC - λσmax was 0.21 nm with the same Tc, and the temperature difference ΔT = TM - TN between the monolithic microchip DFL (TM) and the uncoated Nd:YVO4 crystal (TN) was numerically calculated as 54°C with the same wavelength of λSEC and λσmax. Considering the DFL signals only emitted within the ECS spectral range of the laser crystal, the wavelength of λSEC and λσmax should be identical. Here, the ΔT can be explained as the different pump and laser powers, the heat deposited in the fluorescence spectra acquisition experiments were much smaller than that in the laser spectra acquisition experiments. As a result, the monolithic microchip DFL temperature TM was higher than the uncoated Nd:YVO4 crystal TN which was equal to the heat sink temperature Tc.
Fig. 4 (a) The λSEC and λσmax vs. the heat sink temperature Tc; (b) The laser spectra and the compensated ECS spectra at Tc = −4°C, 16°C, and 36°C; (c) The balanced intensity DFL signals at Tc = −5.6°C and 70.2°C.
By red-shifting the uncoated Nd:YVO4 ECS spectrum 0.21 nm toward the long wavelength direction, the 54°C temperature difference ΔT was compensated. The DFL spectra as well as the compensated ECS spectra with Tc at −4°C, 16°C, 36°C were shown in Fig. 4(b). The results showed that the closer DFL average wavelength‾λ toward the compensated Nd:YVO4 ECS spectral peak wavelength‾λσmax, the better IBR was achieved. When the‾λ located at the long-wavelength side of the‾λσmax, an unbalanced intensity DFL signal with a larger intensity of short-wavelength component appeared, showed in Fig. 4(b) middle. Otherwise, presuming the‾λ located at the short-wavelength side of the‾λσmax, the unbalanced intensity DFL signal with a larger intensity of long-wavelength component was naturally appeared. The DFL signal IBR depended on the relative positions of the‾λ and the‾λσmax.
After carefully controlling the heat sink temperature Tc, when the‾λ shifted to coincident with the‾λσmax, the DFL signals with balanced intensity were achieve at Tc = −5.6°C and 70.2°C, showed in Fig. 4(c). When Tc = −5.6°C, the DFL signal wavelengths were 1064.26 nm and 1064.52 nm, corresponding to the frequency separation of 67.52 GHz. The IBR was 0.991 and the overall output power was measured as 264 mW. When Tc = 70.2°C, the DFL signal had wavelengths of 1064.56 nm and 1064.81 nm, corresponding to the frequency separation at 67.49 GHz. The IBR was 1.009 and the overall output power was measured as 51 mW.
The LD pump power and the as-fabricated Nd:YVO4 microchip thickness (occasionally in sub-wavelength precision) affected Tc for the best IBR. In the experiments, with a higher pump power, a slightly lower specific temperature Tc was needed for the best IBR. Also by theoretical calculation, for a 1.0 mm length microchip, the 4.5 nm increment of the microchip thickness resulted in about 1°C temperature decreasing of the specific temperature Tc for the best IBR.
5. Conclusion
A balanced intensity microchip DFL with 67 GHz frequency separation was presented in this paper. The relative position of the DFL average wavelength‾λ and the compensated Nd:YVO4 crystal ECS spectral peak wavelength‾λσmax determined the IBR of the DFL signal. The IBR was tuned by controlling the heat sink temperature Tc. A 264 mW DFL signals with IBR at 0.991 was experimentally realized when Tc controlled at −5.6°C. A lower Tc of the microchip DFL controlled, a higher power of the balanced intensity DFL signal was achieved.
Funding
Open Foundation of State Key Laboratory of Advanced Optical Communication Systems and Networks (2015GZKF03008); Zhejiang Provincial Research Program of Public Welfare Technology Application (2016C31068); Natural Science Foundation of Zhejiang Province (LY14A040008); National Natural Science Foundation of China (NSFC) (11574068); and supported by Zhejiang Provincial Key Lab of Data Storage and Transmission Technology, Hangzhou Dianzi University.
References and links
1. A. J. Seeds, H. Shams, M. J. Fice, and C. C. Renaud, “TeraHertz photonics for wireless communications,” J. Lightwave Technol. 33(3), 579–587 (2015). [CrossRef]
2. G. Kervella, J. Maxin, M. Faugeron, P. Berger, H. Lanctuit, G. Pillet, L. Morvan, F. van Dijk, and D. Dolfi, “Laser sources for microwave to millimeter-wave applications [Invited],” Photon. Res. 2(4), B70–B79 (2014). [CrossRef]
3. G. Danion, C. Hamel, L. Frein, F. Bondu, G. Loas, and M. Alouini, “Dual frequency laser with two continuously and widely tunable frequencies for optical referencing of GHz to THz beatnotes,” Opt. Express 22(15), 17673–17678 (2014). [CrossRef] [PubMed]
4. M. Vallet, J. Barreaux, M. Romanelli, G. Pillet, J. Thévenin, L. Wang, and M. Brunel, “Lidar-radar velocimetry using a pulse-to-pulse coherent rf-modulated Q-switched laser,” Appl. Opt. 52(22), 5402–5410 (2013). [CrossRef] [PubMed]
5. L. Morvan, D. Dolfi, and J. P. Huignard, “RF photonics for lidar systems optically pre-Amplified dual-frequency lidar-radar,” IEEE LEOS Newslett. 19, 12–13 (2015).
6. C. H. Cheng, L. C. Lin, and F. Y. Lin, “Self-mixing dual-frequency laser Doppler velocimeter,” Opt. Express 22(3), 3600–3610 (2014). [CrossRef] [PubMed]
7. C. H. Cheng, C. W. Lee, T. W. Lin, and F. Y. Lin, “Dual-frequency laser Doppler velocimeter for speckle noise reduction and coherence enhancement,” Opt. Express 20(18), 20255–20265 (2012). [CrossRef] [PubMed]
8. S. De, G. Loas, A. El Amili, M. Alouini, and F. Bretenaker, “Theoretical and experimental analysis of intensity noise correlations in an optically pumped, dual-frequency Nd:YAG laser,” J. Opt. Soc. Am. B 30(11), 2830–2839 (2013). [CrossRef]
9. A. El Amili, G. Loas, S. De, S. Schwartz, G. Feugnet, J. P. Pocholle, F. Bretenaker, and M. Alouini, “Experimental demonstration of a dual-frequency laser free from antiphase noise,” Opt. Lett. 37(23), 4901–4903 (2012). [CrossRef] [PubMed]
10. J. Thévenin, M. Vallet, M. Brunel, H. Gilles, and S. Girard, “Beat-note locking in dual-polarization lasers submitted to frequency-shifted optical feedback,” J. Opt. Soc. Am. B 28(5), 1104–1110 (2011). [CrossRef]
11. Y. Qiao, S. Zheng, H. Chi, X. Jin, and X. Zhang, “Electro-optically tunable microwave source based on composite-cavity microchip laser,” Opt. Express 20(27), 29090–29095 (2012). [CrossRef] [PubMed]
12. P. Zhao, S. Ragam, Y. J. Ding, and I. B. Zotova, “Power scalability and frequency agility of compact terahertz source based on frequency mixing from solid state lasers,” Appl. Phys. Lett. 98(13), 131106 (2011). [CrossRef]
13. P. Zhao, S. Ragam, Y. J. Ding, and I. B. Zotova, “Investigation of terahertz generation from passively Q-switched dual-frequency laser pulses,” Opt. Lett. 36(24), 4818–4820 (2011). [CrossRef] [PubMed]
14. A. Rolland, L. Frein, M. Vallet, M. Brunel, F. Bondu, and T. Merlet, “40-GHz photonic synthesizer using a dual-polarization microlaser,” IEEE Photonics Technol. Lett. 22(23), 1738–1740 (2010). [CrossRef]
15. C. Ren and S. L. Zhang, “Diode-pumped dual-frequency microchip Nd:YAG laser with tunable frequency difference,” J. Phys. D Appl. Phys. 42(15), 155107 (2009). [CrossRef]
16. A. McKay and J. M. Dawes, “Tunable terahertz signals using a helicoidally polarized ceramic microchip laser,” IEEE Photonics Technol. Lett. 21(7), 480–482 (2009). [CrossRef]
17. M. Tani, O. Morikawa, S. J. Matsuura, and M. Hangyo, “Generation of terahertz radiation by photomixing with dual-and multiple-mode lasers,” Semicond. Sci. Technol. 20(7), S151–S163 (2005). [CrossRef]
18. Q. Yang, Y. J. Huo, Y. S. Duan, and Y. Y. Zhan, “Double-longitudinal-mode continuous-wave laser with ultra-large frequency difference used for narrowband terahertz-wave generation,” Acta Opt. Sin. 33(5), 0514002 (2013). [CrossRef]
19. M. Hu, Y. Zheng, J. Cai, G. Zhang, Q. Li, X. Zhou, Y. Wei, and Y. Lu, “CW dual-frequency MOPA laser with frequency separation of 45 GHz,” Opt. Express 23(8), 9881–9889 (2015). [CrossRef] [PubMed]
20. Y. J. Huang, Y. S. Tzeng, C. Y. Tang, S. Y. Chiang, H. C. Liang, and Y. F. Chen, “Efficient high-power terahertz beating in a dual-wavelength synchronously mode-locked laser with dual gain media,” Opt. Lett. 39(6), 1477–1480 (2014). [CrossRef] [PubMed]
21. S. L. Zhang, Y. D. Tan, and Y. Li, “Orthogonally polarized dual frequency lasers and applications in self-sensing metrology,” Meas. Sci. Technol. 21(5), 054016 (2010). [CrossRef]
22. G. Shayeganrad and L. Mashhadi, “Dual-wavelength CW diode-end-pumped a-cut Nd:YVO4 laser at 1064.5 and 1085.5 nm,” Appl. Phys. B 111(2), 189–194 (2013). [CrossRef]
23. B. Wu, P. Jiang, D. Yang, T. Chen, J. Kong, and Y. Shen, “Compact dual-wavelength Nd:GdVO4 laser working at 1063 and 1065 nm,” Opt. Express 17(8), 6004–6009 (2009). [CrossRef] [PubMed]
24. C. L. Sung, H. P. Cheng, C. Y. Lee, C. Y. Cho, H. C. Liang, and Y. F. Chen, “Generation of orthogonally polarized self-mode-locked Nd:YAG lasers with tunable beat frequencies from the thermally induced birefringence,” Opt. Lett. 41(8), 1781–1784 (2016). [CrossRef] [PubMed]
25. M. Hu, R. D. An, H. Zhang, Q. F. Huang, and J. H. Ge, “Experimental investigation of a novel microchip laser producing synchronized dual-frequency laser pulse with an 85 GHz interval,” Laser Phys. Lett. 10(1), 015801 (2013). [CrossRef]
26. X. Délen, F. Balembois, and P. Georges, “Temperature dependence of the emission cross section of Nd:YVO4 around 1064 nm and consequences on laser operation,” J. Opt. Soc. Am. B 28(5), 972–976 (2011). [CrossRef]
27. Y. Sato and T. Taira, “Temperature dependencies of stimulated emission cross section for Nd-doped solid-state laser materials,” Opt. Mater. Express 2(8), 1076–1087 (2012). [CrossRef]
28. T. Yoshino and Y. Kobayashi, “Temperature characteristics and stabilization of orthogonal polarization two-frequency Nd3+:YAG microchip lasers,” Appl. Opt. 38(15), 3266–3270 (1999). [CrossRef] [PubMed]
