In the present work we report on the growth, spectroscopy and laser results of LiLuF4:Pr3+ 1.25% in the melt. Room temperature polarized absorption and emission spectra have been recorded and the decay time of the 3P0 manifold has been measured. Finally efficient room temperature laser emission have been obtained at 522.8 nm, 607.25 nm, 640.17 nm and 721.5 nm under 480 nm pumping by means of an optically pumped semiconductor laser.
©2007 Optical Society of America
In the recent years a lot of interest has grown for the development of all-solid state laser sources emitting in the visible region of the electromagnetic spectrum. The perspective for such devices are linked to the development of a new generation of colour displays, new data storage techniques, holographic techniques, calibration stars for astrophysical experiments, and also in biomedical applications tasks.
For that purpose, one of the most promising candidates between rare earth ions is Pr3+. The electronic configuration of Pr is [Xe](closed shell)+4f2, and the energy level scheme is reported in Fig. 1 for the LiLuF4 crystalline host . Looking to Fig. 1 it is clear that a large number of potential laser transitions are available practically throughout the whole visible range up to the infrared. In the same Figure, we show in blue the pump wavelength, and in green, orange, red, and dark-red the different laser transitions that will be analysed in the next sections. Pr3+-doped materials are also attractive as up-conversion, potential tunable UV, and vacuum ultraviolet VUV solid state laser media [1, 2, 3] as well as mid-infrared applications in low phonon materials .
It is well known that the luminescent properties of rare earth ions depend on the matrix in which these ions are placed: several studies have been performed on oxide hosts [2, 3, 5, 6, 7, 8], but all of them have revealed a too strong non-radiative decay from the 3P0 level, quenching the starting laser level. Continuous wave laser action has been demonstrated in YAlO3 , but the best results have been obtained in fluoride materials [10, 11, 12] in both pulsed and cw regime, because of the low phonon energy. The first laser experiments used Xe flashlamps or Ar+-lasers as pumping sources; recently the development of the solid state technology has greatly enhanced the emission using laser diodes or optically pumped semiconductor lasers (OPS) as pumping sources [13, 14].
In Refs. [10, 11, 12, 13, 14], visible laser action has been demonstrated in LiYF4:Pr3+. In this work we will focus our attention on the isomorphous crystal Pr:LiLuF4 (Pr:LLF). In that anisotropic fluoride host, Pr ions enter the Lu3+ S4 site in the scheelite-like tetragonal structure. The differences with respect to the well known YLF consist in the shorter distance of the S4 sites, leading to a closer position of the Pr ions, but on the opposite LLF possesses phonons with lower cut-off energy, larger emission cross sections and the possibility to grow congruently, resulting in a higher optical quality of the grown crystal. In particular we will report on the growth, emission and absorption spectroscopy and laser action at four different wavelengths. For every laser experiment Findlay-Clay and Caird analysis have been carried out in order to fully characterise the cavity and the crystal quality. To the best of our knowledge the reported slope efficiencies are the highest reached for a Pr-based laser so far.
The crystal growth apparatus consists of a home-made Czochralski furnace with conventional resistive heating; special care has been devoted to the quality of the vacuum system, which has an ultimate pressure limit better than 10-7 mbar, and the growth process was carried out in a high-purity (99.999%) argon atmosphere. The 5N powders, as well as their purification from OH- contamination, have been provided by AC Materials (Tampa, Fl., USA).
The sample was grown using LiF and LuF3 powders as raw material for the host and a proper amount of PrF3 powder has been added for achieving the 1.25% nominal doping in the melt. The growing conditions were 5 RPM for the rotation rate and the pulling rate 1 mm/h, and the temperature of the melt was ≈860 °C. The furnace is also provided with an optical apparatus for automatic diameter control. The size of the crystal was about 12 mm in diameter and 55 mm in length (see Fig. 2). The single crystalline character of the sample was checked using a X-ray Laue technique that allows us to identify the c-crystallographic axis and cut oriented samples since the optical characteristics strongly depend on the position of the optical axes. Samples of two different lengths (4.07 mm and 9.20 mm) have been prepared. The crystal end faces (≈3×3 mm2) were polished for laser quality.
ICP analysis has been performed on the sample in order to ascertain the actual doping concentration. In fact it is well known that, due to the ionic radius mismatch between Pr and Lu, the segregation coefficient is lower than one: in our case it has been measured a 0.19% as actual doping concentration against the 1.25% Pr content in the melt.
1.2. Spectroscopy: experimental set-up and results
The polarized absorption spectra were measured by a CARY 500 spectrophotometer between 240 and 2400 nm with a resolution of 0.1 nm in the UV-VIS wavelength region and 1 nm in the NIR region. In Fig. 3 the result obtained for the σ and π polarization are shown in the region where the 3P2, 3P1+1I6 and 3P0 manifolds peaks lies; in particular the 3P0 absorption as a maximum of 13 cm-1 in the E ∥ c polarization at 479.2 nm with a FWHM of 0.46 nm.
Room temperature polarized emission measurements have been performed between 460 and 740 nm by exciting the samples using the 457.9 nm radiation from a cw Argon laser pumping the 3P1+1I6 manifolds with a pump power as little as 3 mW. For the emission measurements, the laser beam was focused onto the samples with a 10 cm focal length lens. The fluorescence signal was detected perpendicularly to the pump laser direction to avoid pump spurious scattering. The luminescence was chopped and focused by a 75 mm focal length Infrasil lens on the input slit of a Jobin-Yvon monochromator with 320 mm focal length, equipped with a Glan-Thomson polarizer at the entrance in order to perform polarized measurements. For the fluorescence measurements we used a 1200 gr/mm grating blazed at 500 nm, and the resolution was set to 0.13 nm. The signal was filtered by proper filters (λt=460 nm) in order to suppress as much as possible any spurious laser scattering from the sample; the fluorescence was detected by a R1464 Hamamatsu photomultiplier, fed into pre-amplifiers, processed by a lock-in amplifier and subsequently stored on a PC. The acquired spectra were normalized for the optical response of the system using a black-body source at 3000 K.
The results are shown in Fig. 4: as can be seen the spectra are reach in features and several transition lines are present; for sake of clarity only some of them have been labelled. The emission of the Pr:LLF is dominated by the transitions starting from the 3P0 level; in particular the most intense line lies in the blue region of the spectrum at 479.6 nm corresponding to the 3P0→3H4 transition to the ground state. Using the absorption data, the Füchtbauer-Ladenburg formula allows us to calculate the emission cross sections at the lasing wavelength; the results are reported in Tab. 1.
Room temperature decay time behaviour of the starting laser level, i.e. the 3P0 manifold. It has been recorded with an apparatus similar to that described for the steady state measurements, but in this case the pumping source was a frequency doubled pulsed tunable Ti:Al2O3 (pulse duration ≈ 30 ns) laser tuned to the maximum absorption of the 3P2 manifold around 442 nm. The sample was pumped near an edge and the fluorescence was collected from a thin section (≈ 1 mm) of the sample to observe a uniformly pumped volume and reduce radiation trapping effect. Furthermore the power incident on the sample was reduced by means of attenuators to suppress non-linear effects as much as possible; we estimated the incident energy as little as 10 nJ. The signal from the photomultiplier was amplified, by fast amplifiers, and then processed by a digital oscilloscope, connected to a PC for storing purpose.
As can be seen in Fig. 5, the decay time curve exhibits a single exponential behaviour that has been fit with τ = 45.4 ± 0.3 μs. This result is in agreement with that of Ref. , where the authors have measured 48 μs in a 0.026% Pr-doped LLF.
2. Laser setup and results
The pump process was done by a cw frequency-doubled OPS laser (Coherent Lübeck GmbH, Lübeck, Germany), providing a linear polarized beam at 480 nm wavelength with nearly diffraction limited emission (M 2≈1). The wavelength of the OPS was tuned to 479.5 nm for optimal absorption in the Pr:LLF laser crystal. The maximum output power available at this wavelength was 220 mW. This laser was only slightly tunable and it does not allow us to reach the absorption peak lying at 479.2 nm: for this reason 479.5 nm have been chosen as pumping wavelength (it was the shortest reachable wavelength) where the maximum output power was 220 mW. The pump laser was focused on the sample by a 50 mm focal length lens, resulting in a spot size of 28 μm and confocal parameter of 12 mm. Uncoated Pr:LLF samples have been placed in a concentric resonator (cavity length≈90 mm, mirrors radii: 50 mm, confocal parameter: 24 mm) with their c-axis parallel to the pump laser polarization. For all the experiments proper input couplers have been used, having high transmission at the pump wavelength and high reflectivity at the lasing wavelength; in order to perform a detailed analysis of the different laser transitions, several output couplers were used for every experiment. For all the experiments the pumping wavelength and the laser emission have been measured with a Fourier-transform-spectrometer: the spectral stability have been checked repeating several times the wavelength measurements also in different days, resulting always in the same value within the experimental error. Regarding the temporal laser stability, the output power has been monitored resulting in a rms noise of less than 2% over 2 hours.
2.1. 640.2 nm: 3P0→3F2 transition
The first laser transition that has been measured was the “red” one around 640 nm corresponding to the 3P0→3F2 transition. Only for that experiment two different sample thickness have been used in order to determine the optimal crystal length, 4.07 mm and 9.20 mm respectively. In both cases the output wavelength was 640.17 nm and was σ-polarized, in agreement with the intensity shown in Fig. 4. The results for the two samples under investigation are reported in Tab. 2. As it is evident, the best performances have been obtained for the shortest sample for both slope efficiency and threshold power, and the subsequent laser experiments have been performed only for that one. This result is consistent with the absorbtion measurements of the 3P0 level shown in Fig. 3: at λ=479.47 nm (pumping wavelength), the absorption coefficient is α=5.4 cm-1, resulting in 89% absorption for the 4.07 mm sample and in more than 99% for the longer one; furthermore a longer sample gives rise to higher scattering losses.
The analysis of the results for the 4.07 mm sample, reported in Fig. 6, shows that the lowest threshold is equal to 10mW using a T=1%output coupler, the maximum output power 52.7mW and the maximum slope efficiency ηabs=56%. To the best of our knowledge this is the highest slope efficiency reported up to now in the literature for Pr-based crystals. In Fig. 7 the Findlay-Clay and Caird analysis are reported: both methods indicate a value for the round trip losses equal to 0.4%. Furthermore the Caird analysis suggest a maximum theoretical slope efficiency of 65%; this result is not very far from the maximum value shown in Fig. 6 and Tab. 2.
The best performances of the longer samples were 31 mW as lowest threshold value (three-times more than the 4.07 mm sample) and a maximum slope efficiency of 44% and a maximum output power of 38.9 mW. Findlay-Clay and Caird analysis have been done also for the 9.20 mm long crystal: here we do not report the corresponding figures, but the values are in agreement with those of the shortest sample. This first result for the losses value certifies for the high optical quality of the crystal; for comparison in Ref.  the losses value measured for a Pr:YLF sample were 1.5%. The results are reported in Tab. 3 together with those at the other wavelengths; in addition, for every transition under investigation, we report the overlap efficiency (calculated as the ratio between the maximum possible slope obtained by the Caird analysis and the quantum defect of the transition) and the optical conversion efficiency (i.e. the ratio between the maximum output power and the maximum incident power).
2.2. 721.5 nm: 3P0→3F4 transition
The second most intense transition in Pr is the one around 720 nm corresponding to the 3P0→3F4 transition. The output wavelength was 720.5 nm and also in this case was in σ-polarization. The output power characteristics are shown in Fig. 8. The minimum threshold was 12 mW using a T=0.72% output coupler and the maximum output power 50 mW with ηabs=46%.
Also in this case Findlay-Clay and Caird analysis have been performed and the results are reported in Tab. 3. The obtained value for the round trip losses were 0.3% for both methods, in agreement with the results obtained for the 640 nm laser. The Caird analysis indicates a maximum slope efficiency equal to 57%, and also in this case the agreement with the experimental value is satisfactory.
2.3. 607.2 nm: 3P0→3H6 transition
The last transition under analysis originating from the 3P0 level was in the orange region around 607 nm having as final laser level the 3H6 manifold. Fig. 9a shows the output characteristics as a function of the absorbed power: in this case the emission line was at 607.25 nm but it was in π-polarization. In this case the maximum output power was 34.5 mW with a maximum slope efficiency of 31% and also the threshold was twice than the previous cases being 26 mW with T=1.13% as output coupler.
The different behaviour of this transition is also reflected by the results of the Findlay-Clay and Caird analysis: as can be seen from Tab. 3 the two method do not converge to an unique value for the losses. Furthermore the value obtained with the Findlay-Clay analysis is one order of magnitude higher than the previous laser experiments for transition originating always from the 3P0 level. The possible reason for this result are still under investigation: the possible candidates are a residual ground state absorption to the 1D2 manifold or it can be ascribed to an excited state absorption at the laser wavelength starting from the 3F2 or 3F3 levels to the 3P2+1I6 levels. Further measurements are in progress.
2.4. 522.8 nm: 3P1→3H5 transition
The last laser experiment was in the green region around 523 nm corresponding to the 3P1→3H5 transition. The 3P1 manifold is thermically populated from the 3P0 level being the energy difference only 550 cm-1, that can be easily bridged at room temperature (200 cm-1) by one phonon (for LLF ωmaxphon≈460 cm-1 ). The different starting level of the laser transition reflects on the different output power characteristics shown in Fig. 9b; also in this case the π-polarized laser emission is in agreement with the spectroscopic data. Because of the indirect population mechanism of the upper laser level, the threshold power increases significatively to 63 mW also for an output coupler transmissivity of T=0.74%.With this output coupler the maximum power was 7.1 mW with ηabs=10%. This value increases with increasing transmission of the output coupler up to 33% with T=2.35%, the threshold became 94 mW and the output power reached 16 mW.
The different starting laser level has also consequences on the losses and slope efficiencies analysis. In fact Findlay-Clay plot (see Tab. 3) gives 1.3% as value for the round trip losses while the Caird analysis indicates 0.9%. The increased losses value (at least three times that of the 640 nm and 721 nm laser emission) can be ascribed to the shorter lifetime value of the 3P1 manifold.
3. Conclusions and further developments
In conclusion, we reported on the growth of a Pr:LLF crystal and on the spectroscopic analysis we have performed regarding absorption and emission in both steady state and dynamical regime. We demonstrated efficient continuous-wave laser emission at room temperature in a Pr:LLF sample on several transitions in the green, orange, red and near-infrared spectral ranges, with the highest slope efficiencies for all the transitions, to the best of the authors knowledge. The Findlay-Clay and Caird analysis have been performed for every laser experiment: the low value for the round trip losses demonstrates the high optical quality of the grown crystal. The further development will consist in the power scaling of the system and in the possibility to obtain efficient intracavity UV frequency doubling exploiting the resistance to color center formation of the LLF crystal.
The authors wish to acknowledge Mrs.I. Grassini for helping in the preparation of the samples, Mr. H.P. Jenssen and A. Cassanho for helpful discussions, Coherent Lubeck GmbH, Lübeck, Germany, and the Vigoni programme of CRUI-DAAD.
References and links
1. S. Nicolas, E. Decroix, Y. Guyot, M.-F. Joubert, R. Yu. Abdulsabirov, S. L. Korapleva, A. K. Naumov, and V. V. Semashko, “4f2 to 4f5d excited state absorption in Pr3+-doped crystals,” Opt. Mater. 16,233–242 (2001). [CrossRef]
2. G. Özen, O. Forte, and B. Di Bartolo, “Downconversion and upconversion dynamics in Pr-doped Y3Al5O12 crystals,” J. Appl. Phys. 97,013510 (2005). [CrossRef]
3. G. Özen, O. Forte, and B. Di Bartolo, “Upconversion dynamics in Pr-doped YAlO3 and Y3Al5O12 laser crystals,” Opt. Mater. 27,1664–1671 (2005). [CrossRef]
4. U. Hömmerich, E. Brown, P. Amedzake, S. B. Trivedi, and J. M. Zavada, “Mid-infrared (4.6 μm) emission properties of Pr3+ doped KPb2Br5,” J. Appl. Phys. 100,113507 (2006). [CrossRef]
5. N. V. Kuleshov, A. S. Shinkevich, V. G. Shcherbitsky, V.P. Mikhailov, T. Danger, T. Sandrock, and G. Huber, “Luminescence and time-resolved excited state absorption measurements in Pr3+-doped La2Be2O5 and KGd(WO4)2 crystals,” Opt. Mater. 5,111–118 (1996). [CrossRef]
6. N. V. Kuleshov, V. G. Shcherbitsky, A. A. Lagatsky, V. P. Mikhailov, B. I. Minkovb, T. Danger, T. Sandrock, and G. Huber, “Spectroscopy, excited-state absorption and stimulated emission in Pr3+-doped Gd2SiO5 and Y2SiO5 crystals,” J. Lumin. 71,27–35 (1997). [CrossRef]
7. F. Xiong, X. Lin, Z. Luo, Q. Tan, E. Ma, and Y. Huang, “Spectroscopic properties of Pr3+ ions in biaxial LaB3O6 crystal,” J. Appl. Phys. 99,064905 (2006). [CrossRef]
8. X. Lu, Z. You, J. Li, Z. Zhu, G. Jia, B. Wu, and C. Tu, “Optic characteristics of Pr3+ doped NaY(MoO2)2 crystal,” Appl. Phys. B 85,585–589 (2006). [CrossRef]
9. T. Danger, A. Bleckmann, and G. Huber, “Stimulated emission and laser action of Pr3+-doped YA1O3,” Appl. Phys. B 58,413–420 (1994). [CrossRef]
10. L. Esterovitz, R. Allen, M. Kruer, F. Bartoli, L. S. Goldberg, H. P. Jenssen, A. Linz, and V. O. Nicolai, “Blue light emission by a Pr:LiYF4 laser operated at room temperature,” J. Appl. Phys. 48,650–652 (1977). [CrossRef]
11. T. Sandrock, T. Danger, E. Heumann, G. Huber, and B. H. T. Chai, “Efficient Continuous Wave-Laser Emission of Pr3+-Doped Fluorides at Room Temperature,” Appl. Phys. B 58,149–151 (1994). [CrossRef]
12. A. A. Kaminskii, A. I. Lyashenko, N. P. Isaev, V. N. Karlov, V. L. Pavlovich, S. N. Bagaev, A. V. Butashin, and L. E. Li, “Quasi-cw Pr3+:LiYF4 laser with λ=0.6395 μm and an average output power of 2.3 W,” Quantum Electron. 28,187–188 (1998). [CrossRef]
14. A. Richter, N. Pavel, E. Heumann, G. Huber, D. Parisi, A. Toncelli, M. Tonelli, A. Diening, and W. Seelert, “Continuous-wave ultraviolet generation at 320 nm by intracavity frequency doubling of red-emitting Praseodymium lasers,” Opt. Express 14,3282–3287 (2006). [CrossRef] [PubMed]
15. A. Yoshikawa, K. Kamada, M. Nikl, K. Aoki, H. Sato, J. Pejchal, and T. Fukuda, “Growth and luminescent properties of Pr:KY3F10 single crystal,” J. Cryst. Growth 285,445–449 (2005). [CrossRef]