Single crystals of Yb:RbTiOPO4 codoped with Nb5+ or Ta5+ were grown by the top seeded solution growth slow-cooling technique. The ytterbium concentration in the crystals varies as a function of the molar ratio of the precursor oxides and of the codopant, reaching a maximum value of 1.9×1020 Yb3+ ions/cm3. The broad band near 1 μm in absorption and emission spectra at room temperature is due to the large splitting of the Yb3+ ground state. The ytterbium 2F5/2 level radiative lifetime in Nb:RbTiOPO4(τrad = 2.7 ms), was calculated and then compared to the measured fluorescence decay time (τem = 2.2 ms), giving an intrinsic quantum efficiency of 81%. To evaluate the potentiality of these crystals for self-frequency doubling, preliminary results of Yb3+ laser operation and fundamental wavelength measurements for type-II non-critical second harmonic generation (λNCPM) are also reported.
© 2007 Optical Society of America
Solid state laser sources in the visible are of great importance in laser technology because they can be used for several applications such as high density optical storage, laser displays and underwater communications. Furthermore, nonlinear optical crystals that can incorporate active ions represent a class of bifunctional materials where the nonlinear and laser effects take place simultaneously. Up to now the active ion which has been mainly used as a dopant in non-centrosymmetric hosts to obtain self-frequency doubling laser action, was neodymium . Nowadays there is an increasing interest in laser technology to replace neodymium by ytterbium, because both ions operate in the same wavelength range, but ytterbium has a number of advantages mainly related to its simple two-manifold electronic structure. For this purpose non-centrosymmetric hosts such as YCa4O(BO3)2 (YCOB) , GdCa4O(BO3)3 (GdCOB) , MgO:LiNbO3(MgO:LNB)  and YAl3(BO3)4 (YAB)  have been doped with ytterbium and self-frequency doubling laser action in the visible was obtained. The advantages of using ytterbium as doping ion in nonlinear optical materials to obtain a self-frequency doubling laser are that this ion does not exhibit absorption in the green, which means no absorption losses for the generated second harmonic beam, and when broad emission bands exist, some of these materials can produce tunable laser radiation in the visible, extending from the green to the yellow. The broad tunability, when existing, is promising to obtain femtosecond pulses.
RbTiOPO4(RTP) belongs to the KTiOPO4(KTP) family of nonlinear optical crystals which crystallize in the orthorhombic system, with the non-centrosymmetric space group Pna21. RTP, as many of the crystals belonging to the KTP family, is of interest due to its excellent nonlinear and thermo-mechanical properties . Its crystallization region has been previously determined . The variation of the crystallization region in self-flux when introducing niobium and ytterbium , and tantalum and ytterbium  has also been investigated.
In this work we study the variation of ytterbium concentration in single crystals of Yb:Nb:RTP and Yb:Ta:RTP obtained by the Top-Seeded-Solution-Growth Slow-Cooling (TSSG-SC) technique using self-fluxes and tungsten-containing fluxes.
We report optical absorption and emission spectra recorded at room and low temperature for Yb:Nb:RTP and Yb:Ta:RTP single crystals, and the corresponding Stark splitting of the Yb3+ electronic states which are compared.
Finally, we discuss the continuous-wave laser operation achieved together with experimental results obtained for the fundamental wavelength in type-II non-critical phase-matching (λNCPM) along the principal axes x and y for Yb:Nb:RTP, to evaluate the potentiality of these crystals as self-frequency doubling materials.
2. Crystal growth
We grew Yb:Nb:RTP and Yb:Ta:RTP single crystals in self-fluxes and tungsten-containing fluxes by the TSSG-SC method without pulling. All the growth experiments were performed in a vertical tubular single-zone furnace, with the crucible inside it, in a position that ensures an axial temperature gradient in the solution between 0.15 and 0.6 K/mm, with the surface always colder than any other part of the solution. The determination of the saturation temperature (Ts), which should be done prior to the growth process and after homogenization of the solution, was performed by observing the growth/dissolution of a crystal seed in contact with the surface of the solution. All the crystals were grown onto seeds oriented with the c crystallographic direction perpendicular to the surface of the solution. During the crystal growth experiments the crystals were submerged in the solution, rotating at a constant velocity. Details on the crystal growth processes, such as the axial gradient or the cooling ramp applied are given in Table 1. Once the cooling ramp was completed, the crystals were slowly removed from the solution and kept some millimeters above the solution surface while cooling the surface down to 298 K at 15 K/h, to avoid any thermal stress.
We analyzed the growth conditions of Yb:Nb:RTP and Yb:Ta:RTP crystals grown in self-fluxes and tungsten-containing fluxes. Then we were able to evaluate how the conditions of growth, such as the axial gradient or the cooling program used, and the flux composition used in each case affect the quality of the crystals obtained and the actual ytterbium concentration.
The molar ratio of precursor oxides was almost the same for Yb:Nb:RTP and Yb:Ta:RTP crystals grown in self-fluxes. However, it was different for Yb:Nb:RTP crystals grown in tungsten-containing fluxes. Crystals of Yb:Nb:RTP and Yb:Ta:RTP were grown in self-fluxes from solution with a molar composition (Yb2O3:Nb2O5 or Yb2O3:Ta2O5):TiO2:Rb2O:P2O5 = x:y:32.00-(x+y):40.80-27.20 (experiments 1 to 4 listed in Table 1). To grow Yb:Nb:RTP crystals in solutions containing WO3(experiments 5 and 6 listed in Table 1) the solution composition cannot be the same as the previous one used to grow Yb:Nb:RTP crystals from self-fluxes (experiments 1 to 4). This is because when WO3 is present in the solution the RTP crystallization region shifts to Rb2O richer regions and becomes narrower. Previously, we also determined how the values of the distribution coefficients of ytterbium in Nb:RTP  and Ta:RTP  crystals vary as a function of the Rb2O/P2O5 molar ratio and TiO2 concentration in the initial growth solution. So, to obtain the desired orthorhombic and non-centrosymmetric Pna21 phase with the highest ytterbium distribution coefficient, while still remaining far enough from the border between RTP and other neighboring phases, solutions with different molar compositions had to be used in self-fluxes and in tungsten-containing fluxes.
We have noticed that the use of polished and crack free seeds of almost the same composition as the crystal that was going to be grown improved the quality of the obtained crystals. Moreover, the thickness of the seed in the a crystallographic direction should be as large as possible, because when doping RTP with Nb5+ or Ta5+ the crystals show a plate-like habit. The fact that using this kind of seeds is a way to increase the dimension in the aforementioned direction has been already outlined .
The cooling rate and the cooling interval applied affect the size of the crystals obtained. By applying faster cooling rates spurious nucleation took place on the surface of the solution, which means that shorter cooling interval could be applied.
The single crystals obtained were subjected to electron probe microanalysis (EPMA), to determine their chemical composition, as well as the ytterbium distribution coefficient, defined as K Yb = ([Yb]/([Yb]+[Nb or Ta]+[Ti]))crystal/([Yb]/([Yb]+[Nb or Ta]+[Ti]))solution. These values are listed in Table 1. The distribution coefficients obtained, always smaller that unity, can be well understood since ytterbium ions, as other Ln3+ ions in the Nb:RTP crystals,  are supposed to substitute titanium ions in the structure and the ionic radius of Ti4+(r Ti 4+= 0.605 Å) is much smaller than the radius of Yb3+ (r Yb 3+= 0.868 Å) .
As shown in Table 1, the saturation temperature (Ts) tends to increase slightly with increasing TiO2 content in the solution composition. This rule is not accomplished in the last growth experiment, probably due to a slight change in the solution composition induced by an increase of the solvent evaporation during the homogenization process.
Comparing the first two experiments, growth by using Nb5+ as a codopant in self-fluxes and performed with the same solution composition, the difference is related to the cooling interval applied (14.5 and 17 K respectively) and the seed width used in each case. As mentioned before, the seed width determines the dimension along the a crystallographic axis. The crystal dimension in this direction and the crystal weight were increased significantly by using a larger temperature interval in the total cooling program. However, the key of the process seems to be in the second step of the cooling program. By increasing the interval of cooling by 2 K at a cooling rate of 0.05 K/h, it was possible to enlarge the last part of the cooling program by 0.5 K, which represents a total enlargement of the cooling program of 2.5 K before spurious nucleation appeared on the surface of the solution. This ensures a better control of the supersaturation during the growth process which yields a larger crystal. This better control of the supersaturation seems to affect positively the ytterbium incorporation in the crystals since the ytterbium distribution coefficient was higher in the second experiment.
The crystals obtained in the third and fourth growth experiments, which were grown by using Ta5+ as a codopant, had almost the same weight. However, their dimensions were slightly different, due to the different width of the seed in a direction used. Also in this case, also, the introduction of a preliminary step in the cooling program consisting of a 1 K temperature decrease at a rate of 1 K/h and the reduction of the cooling rate in the second step have beneficial effect in obtaining a better quality crystal in a smaller cooling interval. Thereby the concentration of Ta5+ in the crystals is almost doubled although the ytterbium incorporation remains almost unchanged.
Although size and weight of the crystals containing Nb5+ and Ta5+ are not comparable as they were grown in crucibles with different dimensions and with different conditions of seeding, the ytterbium concentration obtained in these four growth experiments does not vary significantly, so both codopants, Nb5+ or Ta5+, have more or less the same effect in increasing the concentration of Yb3+ in RTP crystals and both can be used for this purpose.
Finally, the addition of WO3 in the solution, has a beneficial effect, as it decreases the solution viscosity, which allows to grow the crystals at a faster rate and use a smaller axial thermal gradient in the crucible (note that although large axial thermal gradients in the crucible are needed to stir the solution when the viscosity is high, they shift the growth conditions further away from the thermodynamic equilibrium, and thus, from ideal crystal quality). This addition of WO3 does not improve the ytterbium incorporation in the crystals but it also does not reduce it significantly with respect to crystals grown in self-fluxes.
3. Ytterbium spectroscopy
We studied the polarized optical absorption of the ytterbium ion in Nb:RTP and Ta:RTP samples at room and low temperatures. The absorption spectra, recorded between 850 and 1100 nm, were measured using a Varian Cary 500 spectrophotometer and a Glan-Taylor polarizer. The low temperature spectra were recorded using a Leybold RDK-6-320 closed-cycle helium cryostat. Figure 1 shows polarized absorption spectra recorded at room temperature for Yb:Nb:RTP and Yb:Ta:RTP. The crystallographic axes a, b and c are parallel to the principal optical axes, x, y and z of the dielectric frame, for which nx<ny<nz is fulfilled. Yb3+ (which occupies a C1 symmetry site) has an odd number of electrons in the 4f shell. Hence, polarization selection rules for the electronic transitions are not expected. Thus, the number and position of the peaks should be independent of the light polarization, but the intensity may vary. This was confirmed in the recorded polarized absorption spectra (Fig. 1): The most intense peak is the one located near 970 nm, for E//a and b crystallographic axes, and the one located at ≈ 900 nm for E//c. However, the number of peaks and their position is the same for the three orthogonal polarizations.
Although two different positions of Yb3+ can be found in the structure , and we have shown previously that the structural sites in which lanthanide ions can be found depend on their concentration [11, 13, 14], no additional peaks were observed on the lanthanide spectra due to multiple sites when studied by site selective spectroscopy and only slight differences in the intensity of the spectra were evident .In fact, for the Yb3+ concentration in the samples characterized in this work, we expect Yb3+ to substitute only for Ti4+ at the two corresponding crystallographic positions (both sites with C1 symmetry) in the structure of these crystals .
As shown in Fig. 1, the main difference between the polarized spectra of Yb:Nb:RTP and Yb:Ta:RTP is that the cross section values obtained for E//c (maximum at ≈ 900 nm) and E//b (maximum at ≈ 970 nm) are quite different (0.58×10-20 cm2 and 0.79×10-20 cm2, respectively) for Yb:Nb:RTP, while for Yb:Ta:RTP they are closer (0.77×10-20 cm2 and 0.82×10-20 cm2). Hence, for Yb:Ta:RTP the pump wavelength could be set to ≈ 900 nm or ≈ 970 nm depending on the polarization chosen. The cross section values already published for Yb:Nb:RTP  are in good agreement with the results reported in this study.
In Fig. 2, the evolution of the absorption cross sections from 6 K to 300 K is shown for samples of Yb:Nb:RTP (a) and Yb:Ta:RTP (b). From the spectra recorded at 6 K we accurately determined the energies of the three Stark sublevels of the upper 2F5/2 multiplet. As shown in the inset of Fig. 3 the energies of these Stark sublevels, do not differ significantly when using niobium or tantalum as a codopant.
The unpolarized fluorescence spectra were recorded between 950 and 1100 nm in a 90° geometry with excitation by a 200 mW diode laser at 940 nm modulated at 1 kHz. The fluorescence was dispersed by a double monochromator (Jobin Yvon-Spex HR460), with a focal length of 0.46 m, and detected by a cooled Hamamatsu NIR R5509-72 photomultiplier which was connected to a lock-in amplifier. To perform the low temperature emission measurements at 10 K a closed-cycle helium cryostat (Oxford CCC1104) was used.
From the emission spectra obtained at 10 K the four sublevels of the ground state, 2F7/2, were determined (see Fig. 3 insets). An additional peak appears at room temperature, centered at 1051.0 nm in Nb:RTP and at 1051.1 nm in Ta:RTP. It is attributed to a transition between the (1’) Stark sublevel of the upper multiplet and the (3) Stark sublevel of the lower multiplet. The large splitting of the Yb3+ ground state, 956 cm-1 for Nb:RTP and 953 cm-1 for Ta:RTP, results in low (0.76% and 0.75%, respectively) population of the highest 2F7/2 sublevel at room temperature. Such high energies of the highest sublevel facilitate the laser operation in a quasi-three level scheme, reducing the laser threshold.
The radiative lifetime value for Yb:Nb:RTP was determined from the spontaneous emission probability (Aif) calculated by averaging (over the polarizations) the integrated absorption coefficients and refractive index using the relationship proposed by Weber 
where gf and gi, are the degeneracies of the final and the initial states (4 and 3, respectively), n is the refractive index, N is the ytterbium concentration (which is equal to 1.87×1020 ions/cm3 in this case) and α(V) is the absorption coefficient at frequency v = c/λ. The result is τrad = 2.7 ms. Comparing it with the fluorescence decay time measured previously , τem = 2.2 ms, one obtains an intrinsic quantum efficiency of 81%. The fluorescence decay time is of the same order as the one obtained for GdCOB  and YCOB , but much larger than in YAB , GdAl3(BO3)4(GAB), MgO:LNB and KGd(PO3)4(KGP), as shown in Table 2.
Sellmeier equations are not available for Ta:RTP because the crystals obtained up to now were not large enough to process prism shaped samples. Nevertheless, the radiative lifetime for Yb:Ta:RTP was also calculated using the dispersion relations for the principal refractive indices nx, ny and nz of Nb:RTP. This is justified because the difference in the nx, ny and nz refractive indices measured at 632.8 nm (by a prism coupler) for the samples used to perform the spectroscopic studies was less than 0.01. The value obtained for Yb:Ta:RTP, τrad = 2.6 ms, was only slightly lower.
4. Laser operation
The experimental set-up scheme (Fig. 4) used to study the laser operation of Yb:Nb:RTP was a standard astigmatically compensated Z-shaped cavity. The pump source was a home-made Ti:sapphire laser (972.7 nm, FWHM < 1 nm, max. 1.8 W), with the beam focused by a 6.28 cm lens onto the sample which is inserted under Brewster angle between the folding mirrors. This fixes the laser polarization and the pump polarization was always in the same plane to avoid pump losses. The estimated Gaussian pump waist in the focus was ≈ 30 μm. The ≈ 150 cm long cavity was terminated by a rear plane mirror and a plane output coupler.
We achieved efficient laser operation of Yb:Nb:RTP. The sample used to study the laser operation had dimensions of ≈ 3 × 2.5 × 3 mm3 (a × b × c), and was obtained from the growth experiment #2 (see Table 1). Four of its faces, those normal to the a and b crystallographic directions, were polished.
The main results obtained are summarized in Table 3. Further experimental details and results have been recently published . The gain curves, shown in Fig. 5, calculated by using the room temperature absorption cross sections from Fig. 2a and the emission cross sections obtained previously by the Füchtbauer-Ladenburg method  demonstrate that at low population inversion, the expected laser wavelength would be almost the same for the three different polarizations. This is in good agreement with the experimental laser wavelengths in Table 3. As can be seen from the table, the laser wavelength is almost the same for the three different polarizations (E//a, E//b and E//c) and the two output couplers (Toc = 1 and 3%). The extremely low laser thresholds obtained are related to the large splitting of the ground state.
The tuning behaviour of the laser was also studied by inserting a birefringent filter in the vicinity of the output coupler (M4). The tuning range was almost the same for the three polarizations, extending from 1009 to 1081 nm (the FWHM for E//b is 59 nm). Fig. 6 shows the tuning range measured for E//b. The structure of the tuning curve consists of three peaks, which are related to the transitions from the two lowest sub levels of the upper multiplet (0’ and 1’) to the two highest sublevels of the lower multiplet (2 and 3). This broad tuning range is also related to the large splitting of the ground state. The maximum achievable output power depends not only on the absorption cross section for the corresponding polarization but also on the recycling effect in the quasi-three-level Yb-laser scheme, i.e. on the actual intracavity laser power. Smaller output coupling results in higher intracavity power which compensates the pump bleaching effect increasing the actual crystal absorption. Thus a maximum output power of 154 mW was obtained with Toc = 1%, for propagation along the a and polarization along the b axis, at an absorbed pump power of 386 mW (optical-to-optical efficiency equal to 40 %).
Table 2 shows some of the spectroscopic as well as estimated and measured laser parameters for the 6 non-centrosymmetric Yb-hosts for which lasing has been demonstrated up to now. To compare these hosts with Nb:RTP, three laser parameters have been calculated. These laser parameters are: βmin(the minimum fraction of Yb3+ ions that must be excited so that the gain equals the ground-state absorption at λext -extraction wavelength), Isat (the pump saturation intensity) and Imin (the minimum pump intensity required for transparency to be achieved at the extraction wavelength). The equations used to calculate these parameters are listed below:
where σabs and σem are the absorption and the emission cross section values respectively λext.
Imin should be as low as possible for efficient laser materials doped with Yb 3+, which means emission cross section as high as possible.
5. Fundamental wavelength for non-critical type-II second harmonic generation
To measure the fundamental wavelength for non-critical type-II SHG (λNCPM) along the principal optical axes of the dielectric frame, we used a tunable OPO (Continuum Panther) pumped by a Nd:YAG laser (Continuum SLI-10) emitting 5 ns-long (FWHM) pulses between 410 nm and 2.55 μm at a repetition rate of 10 Hz. The OPO wavelength was controlled by a Chromex 250 SM scanning monochromator. The OPO beam was properly focused onto the sample cut as a cube which was stuck on a goniometric head placed in the center of a goniometric rotation stage. The polished crystal faces were perpendicular to a and b axes, which were pumped at normal incidence. An achromatic half-wave plate placed between the laser source and the sample, provided an adjustment of the polarization of the OPO beam to ensure type-II SHG for propagation in the ab plane of the crystal.
The cube sample used to measure λNCPM values for type-II SHG of Yb:Nb:RTP in the ab plane, was cut from the same crystal that was used to obtain the sample for laser operation. We found λNCPM = 1118.2 nm at θ= 90°; φ = 0°, corresponding to the x or a axis, and λNCPM = 985.4 nm at θ = 90°; φ = 90°, corresponding to the y or b axis. These data suggest that we have type-II SHG phase-matching directions in the ab plane, for wavelength ranging between 985.4 nm and 1118.2 nm. Especially, by choosing the right azimuthal angle φ, type-II SHG phase-matching conditions is possible for a fundamental wavelength matching the laser wavelength at 1050.6 nm which corresponds to the maximum of the laser emission in Yb:Nb:RTP. Moreover, as the laser tuning range achieved is quite broad, possible deviations of φ as a result of the sample preparation should not be detrimental.
We studied the influence of the growth conditions for ytterbium doped RTP crystals codoped with niobium or tantalum using the TSSG-SC technique. The three strategies used to further increase the Yb3+ concentration in RTP crystals, codoping with Nb5+, codoping with Ta5+, and growth in W-containing solution, have been successful for this purpose, reaching in all cases similar Yb3+ concentrations. A maximum ytterbium concentration of ≈ 2×1020 ions/cm3 has been obtained in samples large enough to perform spectroscopic characterization and study laser operation. From the absorption and the emission measurements in Yb:Nb:RTP and Yb:Ta:RTP the Stark splitting of the two Yb3+ electronic states has been determined, and the obtained values are almost the same for the two codopants. The absorption cross section values are in general similar for both hosts, Nb:RTP and Ta:RTP. However, in Yb:Ta:RTP the maximum absorption cross section values obtained for polarizations E//b and E//c are much closer. Then two different wavelengths (972.2 or 903.8 nm) could be used to pump Yb:Ta:RTP depending on the chosen polarization scheme. Moreover, the Yb3+ upper level radiative lifetime in Nb:RTP has been calculated (τrad = 2.7 ms) and compared with the experimental fluorescence decay time (τem = 2.2 ms) measured at room temperature. Laser operation has been obtained for the first time to our knowledge, in an ytterbium doped crystal belonging to the KTP family. The maximum output power obtained at 1050.6 nm was 154 mW, and the maximum slope efficiency exceeded 60%. Moreover, the broad tuning range achieved (from 1009 to 1081 nm) makes Yb:Nb:RTP promising for mode-locking to obtain ultrashort (femtosecond) laser pulses. Finally, as the laser wavelength for the three possible polarizations is around 1050 nm, self-frequency doubling of the laser action should be possible in Yb:Nb:RTP for a propagation in the ab plane at some azimuthal angle φ, leading to green light generation. Future experiments will be devoted to such demonstration.
This work has been supported by the Spanish government under projects MAT-05-06354-C03-02, MAT-04-20471-E and CIT-020400-2005-14, the Catalan government under project 2005SGR658 and through the EU project DT-CRYS, NMP3-CT-2003-505580. A. Peña thanks the Spanish government for the personal funding BES-2003-1694.
References and links
01. A. Brenier, D. Jaque, and A. Majchrowski, “Bi-functional laser and non-linear optical crystals,” Opt. Mater. 28, 310–323 (2006). [CrossRef]
02. D. A. Hammons, J. M. Eichenholz, Q. Ye, B. H. T. Chai, L. Shah, R. E. Peale, M. Richardson, and H. Qiu, “Laser action in Yb:YCOB (Yb:YCa4O(BO3)3),” Opt. Commun. 156, 327–330 (1998). [CrossRef]
03. H. Zhang, X. Meng, P. Wang, L. Zhu, X. Liu, X. Liu, Y. Yang, R. Wang, J. Dawes, J. Piper, S. Zhang, and L. Sun, “Growth of Yb-doped Ca4GdO(BO3)3 crystals and their spectra and laser properties,” J. Cryst. Growth 222, 209–214 (2001). [CrossRef]
04. L. E. Bausá, M. O. Ramírez, and E. Montoya, “Optical performance of Yb3+ in LiNbO3 laser crystal,” Phys. Stat. Sol. A 201, 289–297 (2004). [CrossRef]
05. P. Dekker, J. M. Dawes, J. A. Piper, Y. Liu, and J. Wang, “1.1 W CW self-frequency-doubled diode-pumped Yb:YAl3(BO3)4 laser,” Opt. Commun. 195, 431–436 (2001). [CrossRef]
06. F. C. Zumsteg, J. D. Bierlein, and T. E. Gier, “KxRb1-xTiOPO4: A new nonlinear optical material,” J. Appl. Phys. 47, 4980–4985 (1976). [CrossRef]
07. J. J. Carvajal, V. Nikolov, R. Solé, Jna. Gavaldà, J. Massons, M. Rico, C. Zaldo, M. Aguiló, and F. Díaz, “Enhancement of the erbium concentration in RbTiOPO4 by codoping with niobium,” Chem. Mater. 12, 3171–3180 (2000). [CrossRef]
08. J. J. Carvajal, V. Nikolov, R. Solé, Jna. Gavaldà, J. Massons, M. Aguiló, and F. Díaz, “Crystallization region, crystal growth, and characterization of rubidium titanyl phosphate codoped with niobium and lanthanide ions,” Chem. Mater. 14, 3136–3142 (2002). [CrossRef]
09. A. Peña, J. J. Carvajal, J. Massons, J. Gavaldà, F. Díaz, and M. Aguiló, “Yb:Ta:RbTiOPO4, a new strategy for further increase the lanthanide concentration in crystals of the KTiOPO4 family,” Chem. Mater. (in press).
10. J. J. Carvajal, C. F. Woensdregt, R. Solé, F. Díaz, and M. Aguiló, “Change in the morphology of RbTiOPO4 introduced by the presence of Nb,” Cryst. Growth & Des. 6, 2667–2673 (2006). [CrossRef]
11. J. J. Carvajal, J. L. García-Muñoz, R. Solé, Jna. Gavaldà, J. Massons, X. Solans, F. Díaz, and M. Aguiló, “Charge self-compensation in the nonlinear optical crystals Rb0.855Ti0.955Nb0.045OPO4 and RbTi0.927Nb0.056Er0.017OPO4,“ Chem. Mater. 15, 2338–2345 (2003). [CrossRef]
12. R. D. Shannon, “Revised effective ionic radii and systematic studies of interatomic distances in halides and chalcogenides,” Acta Cryst. A 32, 751–767 (1976). [CrossRef]
13. D. Bravo, A. Martín, J. J. Carvajal, M. Aguiló, F. Díaz, and F. J. López, “Er3+ impurities in KTiOPO4 studied by electron paramagnetic resonance,” J. Phys.- Condens. Matter 18, 6655–6663 (2006). [CrossRef]
14. D. Bravo, A. Martín, J. J. Carvajal, M. Aguiló, F. Díaz, and F. J. López, “An EPR study of Er3+ impurities in RbTiOPO4 single crystals,” Eur. J. Phys. (Submitted)
15. J. J. Carvajal, R. Solé, Jna. Gavaldà, J. Massons, P. Segonds, B. Boulanger, A. Brenier, G. Boulon, J. Zaccaro, M. Aguiló, and F. Díaz, “Spectroscopic and second harmonic generation properties of a new crystal: Yb-doped RbTiOPO4,” Opt. Mater. 26, 313–317 (2004). [CrossRef]
16. M. Weber, “Optical Properties of Yb3+ and Nd3+-Yb3+ energy transfer in YAlO3,” Phys. Rev. B 4, 3153–3159 (1971). [CrossRef]
17. F. Mougel, K. Dardenne, G. Aka, A. Kahn-Harari, and D. Vivien, “Ytterbium-doped Ca4GdO(BO3)3: an efficient infrared laser and self-frequency doubling crystal,” J. Opt. Soc. Am. B 16, 164–172 (1999). [CrossRef]
18. A. Aron, G. Aka, B. Viana, A. Kahn-Harari, D. Vivien, F. Druon, F. Balembois, P. Georges, A. Brun, N. Lenain, and M. Jacquet, “Spectroscopic properties and laser performances of Yb:YCOB and potential of the Yb:LaCOB material,” Opt. Mater. 16, 181–188 (2001). [CrossRef]
19. P. Wang, J. M. Dawes, P. Dekker, D. S. Knowles, J. A. Piper, and B. Lu, “Growth and evaluation of ytterbium-doped yttrium aluminum borate as a potential self-doubling laser crystal,” J. Opt. Soc. Am. B 16, 63–69 (1999). [CrossRef]
20. Z. Zhu, J. Li, A. Brenier, G. Jia, Z. You, X. Lu, B. Wu, and C. Tu, “Growth, spectroscopic and laser properties of Yb3+-doped GdAl3(BO3)4 crystal: a candidate for infrared laser crystal,” Appl. Phys. B 86, 71–75 (2007). [CrossRef]
21. E. Montoya, J. A. Sanz-García, J. Capmany, L. E. Bausà, A. Diening, T. Kellner, and G. Huber, “Continuous wave infrared laser action, self-frequency doubling, and tunability of Yb3+:MgO:LiNbO3,” J. Appl. Phys. 87, 4056–4062 (2000). [CrossRef]
22. I. Parreu, M. C. Pujol, M. Aguiló, F. Díaz, X. Mateos, and V. Petrov, “Growth, spectroscopy and laser operation of Yb:KGd(PO3)4 single crystals,” Opt. Express 15, 2360–2368 (2007). [CrossRef] [PubMed]
23. P. Wang, J. M. Dawes, P. Dekker, and J. A. Piper, “Highly efficient diode-pumped ytterbium-doped yttrium aluminum borate laser,” Opt. Commun. 174, 467–470 (2000). [CrossRef]
24. X. Mateos, V. Petrov, A. Peña, J. J. Carvajal, M. Aguiló, F. Díaz, P. Segonds, and B. Boulanger, “Laser operation of Yb3+ in the acentric RbTiOPO4 codoped with Nb5+,” Opt. Lett. 32, 1929–1931 (2007). [CrossRef] [PubMed]