A broadband green light source was demonstrated using a tandem-poled lithium niobate (TPLN) crystal. The measured wavelength and temperature bandwidth were 6.5 nm and 100°C, respectively, spectral bandwidth was 36 times broader than the periodically poled case. Although the conversion efficiency was smaller than in the periodic case, the TPLN device had a good figure of merit owing to the extremely large bandwidth for wavelength and temperature. The developed broadband green light source exhibited speckle noise approximately one-seventh of that in the conventional approach for a laser projection display.
© 2014 Optical Society of America
The portable laser projection display (LPD) is a promising technology for mobile applications owing to its small size, power efficiency, and always-in-focus characteristic [1, 2]. A full-color LPD requires three primary color lasers: red (R), green (G), and blue (B). Of these, the R and B light sources are commercially available in the form of compact laser diodes (LD). However, green LDs are currently inadequate because of their low emission efficiency and instability of materials . One alternative way to obtain a compact green light source is second-harmonic generation (SHG) using nonlinear crystals. We have demonstrated a diode-pumped Nd:YVO4/KTP microchip green laser [4, 5], based on SHG.
Although the developed compact green light sources are highly efficient, there remain problems such as narrow spectral linewidths compared with R and B LDs. Using SHG linewidths more than twenty times narrower can be produced with greater coherence than LDs but these are also harmful to a projection display because of image speckle noise due to a random interference pattern among the scattered rays from the projection screen. There are several engineering approaches to reducing the speckle noise: inserting optical elements , moving elements , and illumination with a laser emitting several slightly different green-color wavelengths . Among these, inserting and rotating a diffractive optical element has been a common approach [9, 10]. However, those methods naturally reduced the optical power throughput due to the diffusing process.
As an alternative approach, to reduce the speckle noise we considered producing a wide green spectrum in SHG. Among the R, G, and B colors, G may be most sensitive to speckle noise because of the color sensitivity of the human eye. In general, a broad spectral bandwidth requires a short nonlinear crystal, which reduces the output efficiency. The output power of SHG is proportional to the square of the sample length, resulting in a poor trade-off between the bandwidth and the conversion efficiency. Therefore, we designed a novel quasi-phase matching (QPM) structure to accomplish a good trade-off between bandwidth and conversion efficiency. In this work, a tandem-poled lithium niobate (TPLN) structure was fabricated, and efficient green light generation was demonstrated with both large temperature and spectral bandwidth compared with the periodically poled lithium niobate (PPLN). Furthermore, the broadband green light source was applied in an LPD to demonstrate the effect of bandwidth on speckle noise reduction.
2. Design of spectral broadening in QPM SHG
A broadening of the phase-matching bandwidth in QPM SHG with uniform and chirped grating has been theoretically proposed  and experimentally demonstrated  in waveguide structures. QPM waveguides consist of several segments, and each segment has a different QPM grating period. Consequently, the bandwidth of QPM SHG is broadened, and the overall SHG efficiency can be varied by modulating the grating periods. Here, in order to design a broadband QPM SHG device, we adopted an apodized step-chirped QPM grating structure , which is composed of segment structures with an aperiodic and apodized sequence. The aperiodic part consisted of a step-chirped grating period with a fixed duty ratio (R), which is the ratio of the polarization-reversed region to the QPM period. The apodized parts consist of a variable duty ratio at a fixed grating period (Λ). These methods are based on gradually turning on and off the nonlinear coupling between the fundamental and second harmonic waves at the edges of the grating to suppress fluctuation in the efficiency. The aperiodic segments smoothly detune the grating period away from the phase-matching condition, and/or the duty ratio of the apodized segments are shifted away from the 0.5, as shown in Fig. 1.
Similar works have been reported for an optical parametric amplification [14, 15] and broadband wavelength conversion  for IR waves. However, to the best of our knowledge, this is the first practical implementation of the apodization technique at short wavelengths for green light generation based on SHG.
The electric poling of the TPLN sample was conducted using a pair of liquid electrodes contacting the +/−Z faces of a congruently grown LiNbO3 (CLN) substrate. The + Z face was covered with a layer of a corrugated metal/photoresist pattern to facilitate the electric poling process . The samples were designed to have a distribution of QPM period ranging from 6.59 to 6.76 μm for green SHG with an interaction length of 5.5 mm and the thickness of 0.5 mm . As shown in Fig. 2, positive Z surface of the sample was pattern surface for electrical poling.
Although visualization of the tandem structure was not sufficient by simple images, duty ratio and periodicity were varied as shown in Fig. 2.
3. Experimental results and discussion
First, we validated the temperature bandwidth of SHG using the fabricated TPLN sample. A Q-switched Nd:YVO4 laser beam was loosely focused on the TPLN sample, which was placed in a temperature-controlled oven. We measured the SHG intensity by changing the sample temperature. As shown in Fig. 3, the measured temperature bandwidth of SHG was approximately 100 °C, whereas the theoretically expected value was 120 °C. The measured temperature bandwidth of the TPLN sample was approximately 18 times broader than the 5.5 °C of conventional PPLN with the same interaction length of 5.5 mm. The experimental temperature-tuning curve qualitatively agreed with the theoretical results. Even though there ripples are in the bandwidth, the wider temperature acceptance is very important for the practical application of green light sources for LPDs in various working environments. Although we implemented the tandem structure to suppress efficiency fluctuations at the edges of the grating, ripples still exist in both the simulation and experimental results. Flat gain control by the tandem structure in the short wavelength regions was very sensitive to the design parameters, which caused difficulties in an obtaining optimum structure. Additionally, tandem structure via electrical poling was also hard to achieve owing to the precise control of the short pitch grating, which results in more ripples in the experimental result than in the simulation.
Second, to characterize the spectral properties of SHG in the TPLN sample, the second- harmonic intensity was measured through wavelength tuning using an optical parametric oscillator (OPO) pumped by the third harmonic of an Nd:YAG laser , as shown in Fig. 4.
The sample temperature was fixed at 39 °C for the wavelength tuning measurement under the input power of 140 μW with a beam waist diameter of 280 μm (1/e2).
The measured spectral bandwidth of the SHG green light was approximately 6.5 nm, as shown in Fig. 5, while the bandwidth was theoretically expected to be 5.0 nm. The measured spectral bandwidth of TPLN was 36 times broader than the 0.18 nm of PPLN. The green beam quality (see the beam pattern of SHG in Fig. 4) was sufficient for use in an LPD. This is an advantage of laser light over high-power LDs for use in LPDs. In Fig. 5, the wider spectrum agrees well with the theoretical curve. The measured tuning curve has a slightly broader spectrum compared with the theoretical one. One possible reason is that our input laser (the OPO) had a finite linewidth of approximately 1 nm greater than of the simulation.
We concentrated on the correlation between speckle noise and temporal coherence of the illumination light source. Speckle noise on the screen of an LPD is quantitatively estimated by “contrast”. The contrast (C) is defined as the standard deviation of the intensity divided by the mean intensity of the illuminated light on the screen. If the contrast is less than 4%, then the human eye cannot recognize speckle noise on the projection display. The contrast is given by :Eq. (1), we can easily expect that our achieved broadband green light will reduce the speckle contrast to approximately one-sixth that of the conventional light source with SHG using a microchip KTP/Nd:YVO4 or PPLN crystal [4, 5]. To investigate this effect, we installed a laser projection system and measured the speckle noise pattern using several different green light sources (see Fig. 6).
Here, we adopted the standard speckle measurement method to match the human eye  using a CCD camera (VM-11M5, Viewworks®) and printing paper as the screen. To acquire the scattered image patterns from the screen, the integration time of the CCD camera was set to be similar to the human eye as about 50 ms. The reported average pixel area of the human eye is 5.24 μm2, with a focal length of about 22.8 mm at a normal aperture size of 3.2 mm .The average spot size of a speckle (S) can be determined by :Eq. (2) the average speckle size of the human eye is approximately 18 μm2. To make the speckle measurement with the 9 μm camera pixel equivalent to human perception with a 3.9-times smaller pixel, we increased the f-number accordingly from 7.1 to 27.8, resulting in a focal length of 90 mm. For a practical test of our TPLN device for an LPD, we selected a commercially available LD as an input light source. The used pump LD was emitted continuous waves (CW) with the maximum output power of 2.3 W. It had several peaks in different wavelength as shown in Fig. 7 with a different power level. Also spatial beam shape of the pump LD was not uniform with much horizontal and vertical fluctuation as presented in Fig. 7. Because of these reasons the frequency converted SH spectrum in Fig. 7 was narrower than our expectation which was achieved by tunable ns pulse as shown in Fig. 5.
Under these conditions the TPLN sample exhibited very low conversion efficiency about 0.01%/W. To improve green conversion efficiency with larger spectrum we are preparing an Yb-doped fiber laser with the spectrum bandwidth of 13 nm at pulse duration of 0.5 ps.
In the present, to make a comparison between TPLN and PPLN, the Q-switched Nd:YVO4 laser used for Fig. 3 used here. The obtained speckle contrasts and measured speckle patterns at different green light sources are summarized in Table 1. The speckle contrast values shown in Table 1 were obtained from the standard deviation of the intensity divided by the mean intensity, which was observed with the CCD camera using the described setup (see Fig. 6). As shown in Table 1, we achieved a speckle noise on TPLN device that was less than one-seventh of that on the PPLN sample. From the broadband green spectrum (see Fig. 5) results, we expected about six times lower speckle noise on TPLN. An additional suppression of speckle noise occurred through the broadband multi-modes of the IR LD compared to the longitudinal single-mode. A reduction effect from a multiple-wavelength synthetic green laser was reported in Ref. 8. A strong speckle noise with a contrast of 58.0% was observed in the He-Ne laser owing to the monochromatic narrow linewidth under the same experimental conditions. The achieved speckle contrast of 4.1% from our TPLN green light device on an LPD was acceptable to the human eye.
Finally, we measured the green output power as a function of input power, as shown in Fig. 8. The used pump source is a Nd:YVO4 laser at input wavelength of 1064 nm with repetition rate of 10 kHz and a pulse duration of 38 ns. The focused beam diameter was about 200 μm. To measure the conversion efficiency, the temperature of the TPLN sample was fixed at 39 °C while the fundamental wavelength was fixed at 1064 nm.
The conversion efficiency reached 20% at the low input intensity of 8.7 MW/cm2. From this result, the estimated effective nonlinear coefficient, deff, was approximately 5.0 pm/V. At a high input intensity of ~100 MW/cm2, the conversion efficiency was over 40%. The estimated effective nonlinear coefficient was 0.31 times that of the conventional PPLN (~16 pm/V ).
If we want to obtain the same bandwidth of 6.5 nm for green light with PPLN, a very short sample length of (1.0/6.5) = 0.15 mm would be required because the green SHG output bandwidth is 1 nm for a 1-mm sample length. Consequentially, there should be a trade-off between TPLN and PPLN in terms of the SHG intensity. In general, the SHG intensity is proportional to both the square of the sample length and the nonlinear coefficient at a fixed input power with a phase-matching condition. To take into account second harmonic intensity at the same spectral bandwidth, the intensity of TPLN is 131 times that of PPLN. Although there are ripples within the spectral bandwidth the green power fluctuation was within 10% except around 530.0 nm. The figure of merit of TPLN is sufficiently good and the achieved broader green light source with low speckle noise would be very useful in an LPD.
In summary, a novel tandem-poled lithium niobate device was fabricated for broadband green light generation with a variable quasi-phase matching (QPM) period from 6.59 to 6.76μm and a variable duty ratio from 0.5 to 0.1 in an apodized step-chirped QPM grating structure. The obtained spectral bandwidth of 6.5 nm for the green color was sufficiently broad for application in a speck-less laser projection display. Even though the achieved nonlinear coefficient was reduced, it would be sufficient for stable green light generation owing to the much broader temperature bandwidth under variable operating conditions of the projection environment. As a future study, we plan to minimize the ripples in the spectral window to reduce fluctuations in the output power via an improved tandem design and poling technology. Also we are planning to use an Yb-doped fiber laser with wider spectrum than the current pump LD with a pulse mode at 20 MHz repetition rate.
This research was partially supported by the National Research Foundation (NRF) of Korea funded by the Ministry of Education, Science and Technology (R15-2008-006-02001-0) and (No. 2010-0009146) and also by Asian Laser Center Program in GIST, An author, N. E. Yu, was partially supported by the Happy Tech-program of the NRF (No. 2011-0020956). The authors S.-H. Fu, C.-W. Liu, L.-H. Peng, and A. Kung were supported by the National Science Council, of Taiwan (NSC-100-2120-M-002-017-CC1), Prof. M. Cha acknowledges the support of the Korea Research Institute of Standards and Science under the Metrology Research Center-project.
References and links
1. M. Stern, D. Yavid, C. Tan, C. Wittenberg, N. Nambudiri, A. Strat, M. Slutsky, D. Gonzalez, C. DiFazio, R. Smajek, and D. Baldwin, “ Laser Projection Display (LPD): A Miniature, High Resolution Projection Engine,” in Proceedings of the Third Americas Display Engineering and Applications Conference (The Society for Information Display, 2006), pp. 186–188.
2. T. Mizushima, H. Furuya, K. Mizuuchi, T. Yokoyama, A. Morikawa, K. Kasazumi, T. Itoh, A. Kurozuka, K. Yamamoto, S. Kadowaki, and S. Marukawa, “Late-Newspaper: Laser projection display with low electric consumption and wide color gamut by using efficient green SHG laser and new illumination optics,” SID Symp. Digest Tech. Pap. 37, 1656–1659 (2006).
3. T. Miyoshi, S. Masui, T. Okada, T. Yanamoto, T. Kozaki, S.-i. Nagahama, and T. Mukai, “510–515 nm InGaN-based green laser diodes on c-plane GaN substrate,” Appl. Phys. Express 2, 062201 (2009). [CrossRef]
4. C. Jung, B.-A. Yu, K. Lee, Y. L. Lee, N. E. Yu, D.-K. Ko, and J. Lee, “A compact diode-pumped microchip green light source with a built-in thermoelectric element,” Appl. Phys. Express 1, 062005 (2008). [CrossRef]
8. D. V. Kuksenkov, R. V. Roussev, S. Li, W. A. Wood, and C. M. Lynn, “Multiple-wavelength synthetic green laser source for speckle reduction,” Proc. SPIE 7917, 79170B (2011). [CrossRef]
10. H. Furue, A. Terashima, M. Shirao, Y. Koizumi, and M. Ono, “Control of laser speckle noise using liquid crystals,” Jpn. J. Appl. Phys. 50(9S2), 09NE14 (2011). [CrossRef]
11. T. Suhara and H. Nishihara, “Theoretical analysis of waveguide second-harmonic generation phase matched with uniform and chirped gratings,” IEEE J. Quantum Electron. 26(7), 1265–1276 (1990). [CrossRef]
12. K. Mizuuchi, K. Yamamoto, M. Kato, and H. Sato, “Broadening of the phase-matching bandwidth in quasi-phase-matched second-harmonic generation,” IEEE J. Quantum Electron. 30(7), 1596–1604 (1994). [CrossRef]
13. A. Tehranchi and R. Kashyap, “Design of novel unapodized and apodized step-chirped quasi-phase matched gratings for broadband frequency converters based on second-harmonic generation,” J. Lightwave Technol. 26(3), 343–349 (2008). [CrossRef]
14. M. Charbonneau-Lefort, M. M. Fejer, and B. Afeyan, “Tandem chirped quasi-phase-matching grating optical parametric amplifier design for simultaneous group delay and gain control,” Opt. Lett. 30(6), 634–636 (2005). [CrossRef]
15. C. Heese, C. R. Phillips, L. Gallmann, M. M. Fejer, and U. Keller, “Role of apodization in optical parametric amplifiers based on aperiodic quasi-phasematching gratings,” Opt. Express 20(16), 18066–18071 (2012). [CrossRef]
16. T. Umeki, M. Asobe, T. Yanagawa, O. Tadanaga, Y. Nishida, K. Magari, and H. Suzuki, “Broadband wavelength conversion based on apodized χ (2) grating,” J. Opt. Soc. Am. B 26(12), 2315–2322 (2009). [CrossRef]
17. L.-H. Peng, C.-C. Hsu, and A. H. Kung, “Broad multiwavelength second-harmonic generation from two-dimensional χ(2) nonlinear photonic crystals of tetragonal lattice structure,” IEEE J. Sel. Top. Quantum Electron. 10(5), 1142–1148 (2004). [CrossRef]
18. N. E. Yu, J. W. Choi, H. Kang, D.-K. Ko, C.-M. Ho, S.-H. Fu, C.-W. Hsu, C.-Y. Chu, C.-L. Chen, W.-S. Wang, L.-H. Peng, A. H. Kung, H. J. Choi, B. J. Kim, and M. Cha, “Development of efficient broadband green light source by tandem quasi-phase-matched structure,” in CLEO:2012, OSA Technical Digest (online) (Optical Society of America, 2012), paper CF3A.8. [CrossRef]
19. N. E. Yu, “Tunable optical parametric generation and broad band second harmonic generation in nonlinear optical crystals,” Ph.D. thesis, Pusan National University (2002).
20. J. W. Goodman, Speckle Phenomena in Optics: Theory and Applications (Roberts, 2007).
21. S. Roelandt, Y. Meuret, G. Craggs, G. Verschaffelt, P. Janssens, and H. Thienpont, “Standardized speckle measurement method matched to human speckle perception in laser projection systems,” Opt. Express 20(8), 8770–8783 (2012). [CrossRef]
22. I. Shoji, T. Kondo, A. Kitamoto, M. Shirane, and R. Ito, “Absolute scale of second-order nonlinear-optical coefficients,” J. Opt. Soc. Am. B 14(9), 2268–2294 (1997). [CrossRef]