Open Access
Add to BibSonomyAbstract
We report on the micron-luminescent properties of carbon ion implanted optical channel waveguides in the Nd:MgO:LiNbO3 laser crystals. The confocal fluorescence images of the waveguide’s cross section are presented based on the analysis of the spatial variation of the Nd3+ fluorescence properties. We have found that the carbon ion implanted waveguides exhibit hybrid fluorescence properties of both hydrogen and oxygen ion implanted waveguides, which clearly denotes a “boundary” effect of light and heavy ions for waveguide formation in lithium niobate crystals.
©2010 Optical Society of America
1. Introduction
Lithium niobate (LiNbO3) is one of the most widely used ferroelectric crystals in advanced photonics and nonlinear optics owing to its unique combination of many excellent features [1,2]. Rare-earth ion doped LiNbO3 crystals have become promising active media for various optical applications, such as solid state lasers and optical amplifiers [3,4]. By using a few techniques, optical waveguides have been produced in LiNbO3 wafers, allowing numerous applications in integrated optics. These techniques consist of metal ion in-diffusion [5], proton exchange [6], ultrafast/UV laser writing [7,8], optical induction [9], and ion implantation/irradiation [10–17], etc. As one of the most extensively used physical methods for material modification, ion implantation has shown the unique ability for waveguide formation in more than 100 materials, including single crystals, polycrystalline ceramics, glasses, semiconductors and organic materials [11]. Waveguides in pure LiNbO3 crystals have been fabricated by energetic beams of typical light ions, i.e., H or He, as well as some “heavy” ions (in some cases they are also called “medium-mass” ions), such as O, F, Si, P, Cl, or Cu [10]. The incident energetic ions generate crystal defects through two mechanisms, i.e., electronic and nuclear collisions, which are usually considered to be the main reasons of refractive index changes in the irradiated LiNbO3 crystals [10,11]. In fact, recent research works have shown that the construction mechanisms of light and heavy ion implanted LiNbO3 waveguides are different: the electronic damage must be taken into account in heavy ion cases whilst may be ignored in light ion cases [10,12]. However, there is no direct method to determine the boundary of light and heavy ions for waveguide formation in this crystal. More recently, via the investigation of ion implanted Nd, MgO codoped LiNbO3 (Nd:MgO:LiNbO3) waveguides, we have shown that the confocal photoluminescence spectroscopy of the samples has clearly exhibited the difference of fluorescence properties of the light (H) [18] and heavy (O) [19] ion implanted waveguides. The luminescence emission properties were modified mainly in the waveguide-substrate boundary for H implanted waveguides and in the waveguide region for O implanted ones, which are the consequence of the induced damages. The critical factor for these modifications (damages) is the electronic stopping power (Se) of the incident ions, which depends on both the mass and the energy of the incident ions. The electronic damage is strongly dependant on the magnitude of Se and the fluences of the incident ions. It has also been proved that the electronic damage is dominant over the nuclear one when Se is over a certain threshold value, which results in more obvious defect generation in case of heavy ion implantation at the common energy and fluence ranges employed for waveguide formation. In light-ion-implanted waveguides, the nuclear damage plays the main role since the Se is very low. Carbon ions are often utilized to fabricate waveguides in optical materials [10]. Since the particularity of this element (atomic number A = 6), one may expect that the C-implanted LiNbO3 waveguides possess the hybrid features of those produced by either light or heavy ion beams. The purpose of this work is two-sided: firstly to report the properties of the channel waveguides in Nd:MgO:LiNbO3 fabricated by MeV C ion beams and, secondly, to show the hybrid effects of the waveguide fluorescence features, which is helpful to distinguish light and heavy ions for waveguide formation in LiNbO3 crystals.
2. Experimental
The x-cut Nd:MgO:LiNbO3 wafers (doped by 0.2 mol% Nd3+ and 4 mol% MgO) were cut with dimensions of 1.5(x)×7(y)×10(z) mm3 and optically polished. By using the methods performed in previous work [20], the channel waveguides have been produced by C+ ion implantation at energy of 3 MeV and fluence of 7×1014 ions/cm2 through a 1.7 MV tandem accelerator at Peking University. After the implantation the sample is annealed at 260°C for 30-90 min in air to improve the guiding qualities of the waveguides. The characterization of the refractive index changes of the waveguide was performed by using the m-line technique at wavelength of 633 nm. The waveguide loss was measured by the Fabry-Perot method [21].
Micro-luminescence experiments were made by using an Olympus BX-41 confocal microscope in combination with an XY motorized stage. The 10-mW continuous wave 488 nm radiation from an Argon laser was focused 10-µm deep from the sample surface by using an oil immersion 100 × microscope objective with numerical aperture (N.A.) = 1.45. In this configuration the 488-nm laser radiation excites the Nd3+ ions from their ground state (4I9/2) up to the 2G3/2 excited state. Then the subsequent 4F3/2→4I9/2 emission band from Nd3+ ions is back-collected by the same microscope objective and analyzed on a high resolution spectrometer. Three dimensional spectral maps (line intensity, FWHM, and energy) of the main fluorescence line were obtained by fitting the collected spectra and plotting the obtained values with the aid of software LabSpec© and WSMP©.
3. Results and discussion
The extraordinary refractive index (ne) of C implanted waveguides has a typical “well” + “barrier” type distribution. A positive change Δne,w ≈ + 0.009 of the well and a negative variation Δne,b ≈-0.01 have been determined in the C implanted sample surface (well) and the boundary of the waveguide and substrate (barrier). In addition, such waveguides are with propagation loss of ~2.7dB/cm at wavelength of 632.8 nm.
For further understanding of the implantation-induced modifications in the crystal, we calculated the electronic (Se) and nuclear (Sn) stopping powers of the 3 MeV C+ ions based on the SRIM 2008 code [22] (Fig. 1 ). It seems reasonable that the optical barrier built up at the end of C ion range inside the crystal is induced by the nuclear collisions, which is similar to the case of light ion implanted waveguides. It has also been reported that the Se of the incident ions may play a more important role when Se is enough high. According to the previous work, the threshold of electronic damage for the amorphization and pre-amorphization is Se,th(am) ≈5-6 keV/nm [10,12] (related to formation of amorphous tracks for every single ion impact) and Se,th(pre) ≈2.2 keV/nm [10,23] (referring to damage accumulation by several ion impacts), respectively, for the ion irradiated LiNbO3. As one can see, the maximum electronic stopping power Se,max ≈2 keV/nm of the C ions is on the surface of the crystal, which is even less than the Se,th(pre). It is noted that for 500keV H and 3MeV O implanted LiNbO3, the surface Se ≈0.1 keV/nm and 2.7 keV/nm, respectively. Therefore, it is easily understood that for 500keV H ions, the contribution of electronic damage is negligible whilst for 3MeV O ions this factor may play a more important role than that of the 3MeV C ions at comparable fluences.
Fig. 1 Distribution of the electronic (solid line) and nuclear (dashed line) stopping powers of the 3 MeV C+ ions in Nd:MgO:LiNbO3 crystals based on the SRIM 2008 code.
For purpose of practical applications involving optical gain, it is necessary to investigate the micro-photoluminescence properties of the active waveguides. Also according to the recent works [24], the fluorescence features in the modified regions may offer fundamental information concerning the mechanisms of waveguide formation. In particular, it has been demonstrated that Nd3+ ions constitute an outstanding optical probe for the detection of structural modifications taking place at the micrometric and sub-micrometric scales in the LiNbO3 network. This is based on the non-vanishing sensitivity of the Nd3+ fluorescence lines against small changes in the LiNbO3 network (such as defects, anisotropic distortions, disordering and damage), as it has been previously demonstrated [18,19,24]. One of the advantages of this technique is that it can measure the presence of very low defect concentrations in LiNbO3 crystals produced by ion implantation. Figures 2(a) and 2(b) show the comparison of the room-temperature photoluminescence spectra obtained in the 3MeV C implanted Nd:MgO:LiNbO3 channel waveguide (after annealing at 260°C for 30 min) and the bulk, corresponding to the emission bands of 4F3/2→4I9/2 and 4F3/2→4I11/2, respectively. The spectra obtained in the waveguides highly resemble those in the bulk, exhibiting very similar shape and intensity, which, in general, means that the spectroscopic properties of the 4F3/2 metastable state of Nd3+ ions are well preserved in the channel waveguide. The absence of any reliable luminescence reduction at the waveguide’s volume indicates the absence of ion-implantation induced fluorescence quenching. This is an advantageous feature for laser applications since it ensures low threshold laser oscillations.
Fig. 2 The comparison of the room-temperature photoluminescence spectra obtained in the C implanted Nd:MgO:LiNbO3 channel waveguide (after annealing at 260°C for 30 min) and the bulk, corresponding to the emission bands of (a) 4F3/2→4I9/2 and (b) 4F3/2→4I11/2.
In order to obtain more detailed information about the micro-structural modification induced in the LiNbO3 network as a consequence of the ion implantation procedure, we have performed a 2D fluorescence scan over the C implanted waveguide’s cross section. For each position the 4F3/2→4I9/2 fluorescence spectrum [Fig. 3(a) ] has been measured and the main emission line (centered at 11250 cm−1) is fitted in order to extract its intensity, spectral position and width. Figures 3(a)–3(f) depict the obtained spatial distributions of the intensity [(a) and (d)], spectral shift [(b) and (e)] and broadening of the emission line [(c) and (f)] as obtained for the waveguide’s cross section after the annealing at 260°C for 30 min and 90 min, respectively. The dashed lines show the spatial locations of the regions with maximum defect concentrations induced by the nuclear damage. The trapezoid shape of the nuclear damage region (NDR) is owing to the different penetration depth of the incident ions with the wedged implantation mask. According to previous works [18,19], from the fluorescence images obtained in terms of intensity, spectral shift and width the spatial distribution of damage, lattice distortions and disorder can be interfered.
Fig. 3 The spatial distribution of the intensity [(a) and (d)], spectral shift [(b) and (e)] and broadening of the emission line [(c) and (f)] of the cross section for the 3MeV C ion implanted Nd:MgO:LiNbO3 samples after annealing at 260°C for 30 min and 90 min. The dashed lines show the spatial locations of the regions with maximum defect concentrations induced by the nuclear damage. Scale bars are 2 µm, in all the cases.
After the 30-min annealing, the Nd3+ emission intensity is well preserved at the electronic damage region (EDR) and slightly reduced at the NDR for the 3MeV C implanted Nd:MgO:LiNbO3 sample. This indicates that some damage has been produced in the LiNbO3 network as a consequence of the nuclear collisions that take place at the NDR. This damage leads to a higher density of defects and is accompanied by a reduction in the collected Nd3+ fluorescence because these defects act as luminescence killers, and/or because they constitute scattering centers for the fluorescence radiation. The intensity and extension of the observed fluorescence reduction has been found to be reduced after the longer-time annealing (90 min), suggesting partial thermal recombination of the damage defects created at the NDR.
As it has been discussed in prior works [18,19], any shift in the Nd3+ fluorescence lines indicates the modification in the crystal field acting on the neodymium ions. This crystal field modification, in turns, reveals an alteration in the distances between Nd3+ ions and its neighbors, i.e., a lattice distortion. As can be observed, for the sample annealed for 30 min, the spectral shift is relevant both at the EDR (waveguide) and at the NDR (optical barrier) [Fig. 3(b)]. This suggests that the LiNbO3 network has been originally distorted in both the EDR and NDR. This hypothesis is further supported by the spectral width fluorescence map [Fig. 3(c)]. The fluorescence maps can be now compared to those obtained in light (H) and heavy (O) ion implanted waveguides. Since, besides the ion mass, the fluence of the implanted ions also plays an important role for waveguide formation and may influence the luminescence features considerably, the comparison on the fluorescence modifications in the H, C and O implanted waveguides are only reasonable when we limit it within cases that the implanted ions generate comparable nuclear damages in LiNbO3 at certain fluences; and also the selected ion implantation conditions are suitable for construction efficient waveguides in the crystals. For LiNbO3 samples with comparable nuclear damages, we have observed that, in the H implanted Nd:MgO:LiNbO3 waveguides, the fluorescence maps have revealed that the modifications induced in the LiNbO3 network are well located at the NDR, being unperturbed at the EDR [18]. In addition, in the case of O ion implanted waveguides fluorescence images denoted that the LiNbO3 lattice was distorted only at the EDR, being unperturbed at the NDR [19]. Thus, the present case related to C ion implantation shows hybrid luminescence features of light and heavy ion implanted waveguides. This fact seems to be reasonable since the mass of C ions falls between that of H and O ions when the ion fluences are chosen to obtain comparable nuclear damages in the substrates, so that both electronic and nuclear damages are taking place. Very interestingly, we have found that after further thermal treatment (at 260°C for 90 min), the induced LiNbO3 lattice distortions (denoted by the fluorescence shift) in the EDR have been practically removed, which results in a strong similarity with those of the H implanted waveguides. This phenomenon can be explained in terms of a thermal induced elastic relaxation of the LiNbO3 distortions induced at the EDR. Even after this second annealing, the spatial distribution of the LiNbO3 distortions in the C implanted waveguide seems to constitute an intermediate step between H and O implantation. This is more evident in Figs. 4(a) –4(c), where the fluorescence maps obtained in terms of the spectral shift induced in Nd3+ fluorescence for the H, C and O implanted Nd:MgO:LiNbO3 channel waveguides (after certain annealing) are shown, respectively. As can be observed, the induced lattice distortions translate from the NDR to the EDR when the ion mass is increased [from light (H) to heavy (O)]. For the intermediate ion mass (C) the maximum LiNbO3 distortions are located slightly before the NDR, i.e., between the pure EDR and the pure NDR. Thus we can conclude that the ion mass determine the dominant damage mechanism being the nuclear one for light ions such as H, the electronic one for heavy ions such as O and an interplay between both for intermediate mass ions such as C.
Fig. 4 The fluorescence maps obtained in terms of the spectral shift induced in Nd3+ fluorescence for the (a) H (at energy 500 keV and fluence of 6 × 1016 ions/cm2, after annealing at 400°C for 1h), (b) C (at energy 3 MeV and fluence of 7 × 1014 ions/cm2, after annealing at 260°C for 90 min) and (c) O (at energy 3 MeV and fluence of 6 × 1014 ions/cm2, after annealing at 260°C for 90 min) ion implanted channel waveguides in Nd:MgO:LiNbO3 crystals. The dashed lines show the spatial locations of the regions with maximum defect concentrations induced by the nuclear damage.
4. Summary
We report on the luminescence features of the C ion implanted channel waveguides in the laser crystal Nd:MgO:LiNbO3. We have provided experimental evidence of an absence of any deterioration in the fluorescence properties of Nd ions, this making the fabricated structures good candidates for the development of nonlinear integrated laser sources. The fluorescence images of the C implanted waveguides have revealed an intermediate situation between the micro-structural modifications caused by light and heavy ion implantations. Our results also show that, by using the confocal luminescence technique, it is possible to locate very low defect concentrations induced in the ion implanted LiNbO3 crystals.
Acknowledgments
The authors gratefully acknowledge the financial supports from the National Natural Science Foundation of China (Nos. 10925524 and 10875075), 973 program (No. 2010CB832906), the Spanish MICINN (No. MAT2007-64686), and UAM-CM (No. CCG08-UAM/MAT-4434).
References and links
1. L. Arizmendi, “Photonic applications of lithium niobate crystals,” Phys. Status Solidi, A Appl. Res. 201(2), 253–283 (2004). [CrossRef]
2. K. K. Wong, “Properties of Lithium Niobate,” (INSPEC, London, 2002).
3. M. N. Palatnikov, I. V. Biryukova, N. V. Sidorov, A. V. Denisov, V. T. Kalinnikov, P. G. R. Smith, and V. Y. Shur, “Growth and concentration dependencies of rare-earth doped lithium niobate single crystals,” J. Cryst. Growth 291(2), 390–397 (2006). [CrossRef]
4. T. Y. Fan, A. Cordova-Plaza, M. J. F. Digonnet, R. L. Byer, and H. J. Shaw, “Nd:MgO:LiNbO3 spectroscopy and laser devices,” J. Opt. Soc. Am. B 3(1), 140–148 (1986). [CrossRef]
5. I. P. Kaminow and J. R. Carruthers, “Optical waveguiding layers in LiNbO3 and LiTaO3,” Appl. Phys. Lett. 22(7), 326–328 (1973). [CrossRef]
6. E. M. Rodríguez, D. Jaque, E. Cantelar, F. Cussó, G. Lifante, A. C. Busacca, A. Cino, and S. R. Sanseverino, “Time resolved confocal luminescence investigations on Reverse Proton Exchange Nd:LiNbO3 channel waveguides,” Opt. Express 15(14), 8805–8811 (2007). [CrossRef] [PubMed]
7. S. Mailis, C. Riziotis, I. T. Wellington, P. G. R. Smith, C. B. E. Gawith, and R. W. Eason, “Direct ultraviolet writing of channel waveguides in congruent lithium niobate single crystals,” Opt. Lett. 28(16), 1433–1435 (2003). [CrossRef] [PubMed]
8. R. R. Thomson, S. Campbell, I. J. Blewett, A. K. Kar, and D. T. Reid, “Optical waveguide fabrication in z-cut lithium niobate (LiNbO3) using femtosecond pulses in the low repetition rate regime,” Appl. Phys. Lett. 88(11), 111109 (2006). [CrossRef]
9. P. Zhang, Y. Ma, J. Zhao, D. Yang, and H. Xu, “One-dimensional spatial dark soliton-induced channel waveguides in lithium niobate crystal,” Appl. Opt. 45(10), 2273–2278 (2006). [CrossRef] [PubMed]
10. F. Chen, “Photonic guiding structures in lithium niobate crystals produced by energetic ion beams,” J. Appl. Phys. 106(8), 081101 (2009). [CrossRef]
11. P. D. Townsend, P. J. Chandler, and L. Zhang, “Optical Effects of Ion Implantation,” (Cambridge Univ. Press, Cambridge, 1994).
12. A. Rivera, J. Olivares, G. García, J. M. Cabrera, F. Agulló-Rueda, and F. Agulló-López, “Giant enhancement of material damage associated to electric excitation during ion irradiation: The case of LiNbO3,” Phys. Stat. Solidi A 206(6), 1109–1116 (2009). [CrossRef]
13. S. M. Kostritskii and P. Moretti, “Specific behavior of refractive indices in low-dose He+-implanted LiNbO3 waveguides,” J. Appl. Phys. 101(9), 094109 (2007). [CrossRef]
14. G. G. Bentini, M. Bianconi, M. Chiarini, L. Correra, C. Sada, P. Mazzoldi, N. Argiolas, M. Bazzan, and R. Guzzi, “Effect of low dose high energy O3+ implantation on refractive index and linear electro-optic properties in X-cut LiNbO3: Planar optical waveguide formation and characterization,” J. Appl. Phys. 92(11), 6477–6483 (2002). [CrossRef]
15. J. Olivares, G. García, A. García-Navarro, F. Agulló-López, O. Caballero, and A. García-Cabañes, “Generation of high-confinement step-like waveguides in LiNbO3 by swift heavy ion-beam irradiation,” Appl. Phys. Lett. 86(18), 183501 (2005). [CrossRef]
16. H. Hu, F. Lu, F. Chen, B. R. Shi, K. M. Wang, and D. Y. Shen, “Monomode optical waveguide in lithium niobate formed by MeV Si+ ion implantation,” J. Appl. Phys. 89(9), 5224–5226 (2001). [CrossRef]
17. M. Bianconi, G. G. Bentini, M. Chiarini, P. De Nicola, G. B. Montanari, A. Nubile, and S. Sugliani, “Defect engineering and micromachining of Lithium Niobate by ion implantation,” Nucl. Instrum. Methods Phys. Res. B 267(17), 2839–2845 (2009). [CrossRef]
18. D. Jaque and F. Chen, “High resolution fluorescence imaging of damage regions in H+ ion implanted Nd:MgO:LiNbO3 channel waveguides,” Appl. Phys. Lett. 94(1), 011109 (2009). [CrossRef]
19. D. Jaque, F. Chen, and Y. Tan, “Scanning confocal fluorescence imaging and micro-Raman investigations of oxygen implanted channel waveguides in Nd:MgO:LiNbO3,” Appl. Phys. Lett. 92(16), 161908 (2008). [CrossRef]
20. F. Chen, “Construction of Two-Dimensional Waveguides in Insulating Optical Materials by Means of Ion Beam Implantation for Photonic Applications: Fabrication Methods and Research Progress,” Crit. Rev. Solid State Mater. Sci. 33(3), 165–182 (2008). [CrossRef]
21. R. Regener and W. Sohler, “Loss in Low-Finesse Ti:LiNbO3 Optical Waveguide Resonators,” Appl. Phys. B 36(3), 143–147 (1985). [CrossRef]
22. J. F. Ziegler, computer code, SRIM http://www.srim.org.
23. G. G. Bentini, M. Bianconi, L. Correra, M. Chiarini, P. Mazzoldi, C. Sada, N. Argiolas, M. Bazzan, and R. Guzzi, “Damage effects produced in the near-surface region of x-cut LiNbO3 by low dose, high energy implantation of nitrogen, oxygen, and fluorine ions,” J. Appl. Phys. 96(1), 242–247 (2004). [CrossRef]
24. E. M. Rodríguez, D. Jaque, E. Cantelar, F. Cussó, G. Lifante, A. C. Busacca, A. C. Cino, and S. R. Sanseverino, “Time resolved confocal luminescence investigations on Reverse Proton Exchange Nd:LiNbO(3) channel waveguides,” Opt. Express 15(14), 8805–8811 (2007). [CrossRef] [PubMed]
