Open Access
Add to BibSonomyAbstract
The spectroscopic properties of ceramic Er3+:Y2O3 relevant to the 2.7 µm mid-IR laser transition were studied in the temperature range of 77–300K. We present the results of experimental measurements of absorption and fluorescence which were used to determine the branching ratios and fluorescence quantum efficiency of the 4I11/2 upper laser level. We have shown that the quantum yield of the mid-IR transition more than doubles when the sample is cooled from room to cryogenic temperature, and the stimulated emission cross-section of the highest peak increases by a factor of 10 for the same temperature change. Updated cryogenic Er3+:Y2O3 laser parameters for the mid-IR ~2.7um transition are presented as well.
© 2016 Optical Society of America
1. Introduction
The strong absorption by water and OH-radicals in the 2.0-3.0 µm mid-IR spectral region makes high-power lasers operating there ideal for such applications as remote atmospheric sensing [1], medical procedures [2], and wind lidar [3]. Such lasers are also useful in molecular spectroscopy, direct infrared countermeasures (DIRCM), and optical pumping of longer wavelength solid state lasers. Among the potential sources for emission in this spectral region, the most developed are the ~2.7-3.0 µm Erbium (Er3+)-doped lasers wherein the desired emission is derived from electronic transitions between the second excited state (4I11/2) and the first excited state (4I13/2). Direct pumping of the upper laser level is easily achieved using GaAs laser diodes operating at ~0.98 µm, and quantum-defect-limited optical-to-optical slope efficiencies of ~30% can be expected [4]. Mid-IR Erbium-based lasers have been demonstrated in CW [5], Q-switched [6], and even efficient cascaded (1.6µm/3.0µm [7]) laser regimes. One drawback to the otherwise promising 4I15/2 → 4I11/2 → 4I13/2 pump-lase scheme is its high quantum defect of ~70% which guarantees significant heat deposition, especially during laser power scaling. For this reason, laser host materials with high thermal conductivity are preferable when designing lasers of this kind.
In recent years, cubic sesquioxides have emerged as a promising alternative to established laser host materials (like garnets) due to their superior thermal conductivities [8] and low maximum phonon energies. The relatively high melting temperature of the sesquioxides (~2450°C [9]) is an obstacle to conventional crystal growth techniques, necessitating a shift in focus to ceramics. Ceramic sesquioxides offer the same beneficial qualities as single crystals at only a fraction of the melting temperature. Among the sesquioxides, we have identified Er3+:Y2O3 as the best gain medium for mid-IR laser operation, because it offers the lowest maximum phonon energy with Ωmax = 597 cm−1 [9]. Low maximum phonon energy is crucial for minimizing the upper laser level multi-phonon (non-radiative) decay rate, and thus maximizing the 4I11/2 radiative lifetime in mid-IR lasers operating around 3µm. In our previous work, we have compared the 4I11/2 fluorescence lifetimes of Er-doped ceramic Y2O3, ceramic Sc2O3, and single-crystal YAG; and the Y2O3 material demonstrated lifetimes nearly an order of magnitude longer than the next best host [10]. Furthermore, temperature dependence of this lifetime measured for Er3+(0.5at.%):Y2O3 ceramic showed an increase from 2.4 ms to 4.2 ms when cooling the sample from room to liquid nitrogen temperature. Due to this increased lifetime and the significantly higher emission cross section at liquid nitrogen temperature, nearly QD-limited 24% optical-to-optical slope efficiency and ~14 W of CW ~2.7µm output power were achieved in a cryogenically-cooled Er3+:Y2O3 ceramic laser [10]. Additionally, this output was strictly pump power-limited.
Since our previous work documenting spectroscopy of the 2 at.% material [11], we have expanded our spectral analysis to ceramic Y2O3 samples with Er3+ concentrations ranging from 0.25 at.% to 10 at.%. One important result of these studies was a clarification of the extent to which radiative reabsorption affects the spectral shape of the 4I11/2 fluorescence as a function of Erbium concentration in this material. As will be shown in the course of this paper, concentrations of 2 at.% already show a significantly altered spectral shape, necessitating us to redo our spectral analysis on a lower concentration sample. For this new analysis, we focused on a ceramic Er3+:Y2O3 sample with a dopant concentration of 0.25 at.%, which exhibited minimal detrimental reabsorption effects. From measurements of this sample as a function of temperature, we were ultimately able to quantify the stimulated emission cross section for the ~2.7µm transition, and determine how it varies from room temperature down to liquid nitrogen temperature. Because the laser transition originates from a level where multiple transitions are possible, the calculation of the stimulated emission cross section requires knowledge of the quantum efficiency and branching ratios from that level.
2. Experimental details
Spectral analysis was performed on several Er:Y2O3 ceramic samples obtained from Konoshima Chemical Co., Ltd. (Japan). These samples were doped at Erbium concentrations ranging from 0.25 at.% to 10 at.% and their thicknesses were all between 3 and 4 mm. The inspection surface of all samples was polished to optical-grade and therefore surface scattering of the laser excitation is negligible.
Absorption measurements corresponding to the ~1 um 4I15/2 → 4I11/2 transition were taken with a Varian Cary 6000i (UV-vis-nIR) spectrophotometer with a resolution of 0.1 nm. Fluorescence spectra from the 4I11/2 to the 4I13/2 (~2.7µm) and 4I15/2 (~1µm) states were obtained using an Acton SpectraPro SP-2750 monochromator equipped with a liquid-nitrogen-cooled InAs photodiode for detection. Excitation into the 4I9/2 manifold was achieved using a Spectra Physics Tsunami Ti:sapphire laser operated in CW mode at 804.2 nm. The fluorescence detection bandwidth was set to 0.1 nm. Fluorescence lifetime measurements of the 4I11/2 manifold were acquired by exciting the sample with an 808 nm laser diode modulated at 10 Hz with a duty cycle of 50%. The decay signal was detected using a filtered Si-like photo diode, and was analyzed using a Tektronix TDS7104 digital oscilloscope. For all spectroscopic measurements, the samples were mounted in a CTI Cryodyne cryogenic refrigerator with temperature tunability between 8K and 300K (room temperature).
The experimental setup for the cryogenic Er:Y2O3 laser is shown in Fig. 1. A simple optical cavity consisted of a 4x10x30mm3 2% Er:Y2O3 sample, which was attached to the standard liquid nitrogen Dewar and cooled from one side. A set of coupling mirrors with reflectivities 60-98% at 2.74 µm have been used in the laser cavity. A Newport model 1918–C power monitor and Thorlabs OSA 205A optical spectrum analyzer were used for power and spectral measurements, respectively. The laser was pumped with spectrally narrowed 974 nm fiber-coupled diode laser acquired from Alfalight Inc.
Fig. 1 Experimental setup for cryogenic Er:Y2O3 laser. F1-F2: collimating lens, RM: rear mirror, CM: coupling mirror, BS: beam splitter, and T1: temperature sensor.
3. Experimental results
In the spectroscopic analysis of the 2.7µm laser transition in Er3+:Y2O3, only the three lowest manifolds of the Erbium ion are relevant, as shown in Fig. 2. This study involved absorption measurements from the ground state to the 4I11/2 manifold, fluorescence lifetimes of this manifold, and fluorescence spectra from 4I11/2 down to the two lower manifolds. From these basic measurements, performed across a wide range of temperatures, vital laser parameters were determined including quantum efficiency and stimulated emission cross sections.
3.1 Absorption
Ground state absorption spectra into the 4I11/2 manifold were measured for several temperatures in the range of 77-300K for the 0.25 at.% Er3+:Y2O3 ceramic sample with a thickness of 3.2 mm. These absorption spectra served as the basis for the majority of the following laser parameter calculations. Room temperature and 77K absorption cross sections were calculated from the absorption spectra using Beer’s law and the rare earth dopant concentration. These cross sections, presented in Fig. 3, show that the highest absorption peak (~980nm) increases by a factor of 4.5 when cooling from room temperature to 77K. The absorption cross section of the laser pump wavelength ~974nm increases by a factor of 3.5 for the same change in temperature. Additionally, the spectral line width of the absorption peaks decreases by 50% when cooling from 300K to 77K.
3.2 Reabsorption
Prior to the current work, fluorescence spectroscopy relevant to the 2.7µm transition had been studied on a 2 at.% Er3+:Y2O3 ceramic sample [11]. Since that time, significant effort has been devoted to determining the effect that Erbium concentration has on the spectral characteristics. Of particular interest was the effect that concentration has on reabsorption. Reabsorption occurs when photons emitted by the fluorescing ion are absorbed by a neighboring ion before they have a chance to escape the material. This reabsorption effect changes the spectral shape of the fluorescence by lowering the intensity in regions where absorption is high.
Shown in Fig. 4(a) is a comparison of the ~980nm fluorescence spectra for a low concentration (0.25 at.% Er, thickness = 3.2 mm) and a high concentration (10 at.% Er, thickness = 4 mm) sample. The inset to Fig. 4(b) shows how the experimental setup was arranged to compensate for the slightly different thicknesses of the samples. Essentially, the excitation beam was focused slightly (< 1 mm) into the sample and the fluorescence collection optics were arranged to sample only a small volume of the excited ions. The sample holder was kept rigid throughout the measurements so that fluorescence was collected from a similar volume for each different sample, thereby ensuring that the observed reabsorption differences depend only on sample concentration. The spectra in Fig. 4(a) are normalized at the ~1030 nm peak because absorption (and therefore reabsorption) is negligible there. When normalizing in such a way, it is immediately apparent the extent to which reabsorption affects the intensity of the fluorescence peaks which occur in areas of high absorption. Most peaks short of the 1030 nm peak exhibit some degree of lowered intensity in the higher concentration sample. The most extreme effect occurs at the highest fluorescence peak at ~980 nm which shows a 50% decrease in peak intensity when comparing the 10 at.% sample to the 0.25 at.% sample. This is to be expected considering this is also the location of the highest absorption peak. As a gauge of the reabsorption effect, Fig. 4(b) shows the 980 nm/1030 nm peak intensity ratio for the entire measured concentration range of samples.
Fig. 4 (a) Fluorescence spectra of the 4I11/2 to ground state transition for Er3+:Y2O3 doped at 0.25 at.% and 10 at.% normalized at the 1030nm peak, and (b) Erbium concentration dependences of the 980nm/1030nm peak intensity ratio and (inset) a diagram illustrating the excitation/emission geometry of the fluorescence measurements.
These reabsorption results are shown to illustrate how important it is to perform spectroscopy measurements on low concentration samples. The results presented in Fig. 4 show how earlier spectral measurements performed on the 2 at.% sample [11] were wildly inaccurate due to reabsorption effects corrupting the fluorescence. For this reason, the remainder of this paper will focus strictly on results garnered from the 0.25 at.% sample.
3.3 Fluorescence lifetimes
Previous measurements of the 4I11/2 fluorescence lifetime as a function of temperature were performed on a Er3+(0.5 at.%):Y2O3 ceramic sample [11]. For the sake of consistency between all measurements in this work, new lifetime data was recorded from the 0.25 at.% sample, and those values are presented in Fig. 5. The lifetime results across the whole temperature range exhibit a single exponential decay as shown by the representative 77K and 300K decay curves in the inset of Fig. 5. The fluorescence lifetime of this sample shows an increase from 2.5 ms to 4 ms when going from room temperature to 77K. Additionally, while at the higher temperatures the lifetime increases linearly, there appears to be a leveling off at the lowest measured temperatures.
Fig. 5 Fluorescence lifetime of the 4I11/2 manifold of 0.25 at.% Er3+:Y2O3 as a function of temperature (K). Inset shows the normalized natural log of the 77 K and 300K PL decay curves.
One possible explanation for this lifetime behavior could be the freezing of phonons at lower temperatures. As lattice vibrations decrease, their contribution to non-radiative decay would lessen which in turn would increase the fluorescence lifetime. Another possible explanation could involve the changing populations among individual Stark levels at different temperatures. Differences in the inter-Stark transition probability between the 4I11/2 and 4I13/2 manifolds could lead to the observed behavior if transitions between the lower-lying Stark levels are less allowed than transitions between higher-lying levels. At this point, we are unable to definitively say which explanation is the cause for this trend. Future measurements of the Raman spectrum as a function of temperature and the inter-Stark transition probabilities will shed light on this matter.
3.4 Fluorescence spectra
Fluorescence spectra for both transitions originating at the 4I11/2 manifold were obtained for temperatures ranging from 77K to 300K. The transition to the 4I13/2 manifold gives the fluorescence centered at 2.7 µm while the transition to the 4I15/2 manifold gives the fluorescence centered at 980 nm. Since the 0.25 at.% Erbium sample was used for these measurements, the reabsorption effect is minimized. Figure 6 shows a representative spectrum including both of these spectral regions measured at 77K. Great care was taken during these fluorescence measurements to ensure that these two spectral regions are scaled appropriately with respect to each other.
Fig. 6 Fluorescence spectrum of both transitions from the 4I11/2 manifold of 0.25 at.% Er3+:Y2O3 at 77 K.
4. Calculations
When there are multiple transitions possible from a given energy level, as there are for the 4I11/2 state of Er3+, careful accounting of the number of photons involved in each transition is necessary when determining the stimulated emission cross sections. The ratio of the number of photons involved in a particular transition to the total number of photons in all transitions from a given upper level is characterized by branching ratio β. In the case of the 4I11/2 manifold of Er3+, the two possible transitions (shown in Fig. 2) are those leading to the 2.7µm fluorescence and the 1.0 µm fluorescence, which would have branching ratios β32 and β31, respectively. Naturally, the branching ratios obey the relationship that β32 + β31 = 1.
Figure 7(a) shows the results of branching ratio calculations for the 4I11/2 to 4I13/2 transition (β32) based on careful analysis of the fluorescence spectra of the two relevant transitions. Due to the complicated nature of analyzing these spectra, error is estimated at approximately 10%. It is shown that at room temperature, 10.5% of the photons are involved in the transition to 4I13/2 leading to the 2.7µm transition. This value agrees well with previous Judd-Ofelt predictions [13,14]. As temperature is decreased to 77K, β32 increases to 17% indicating that more photons are participating in the mid-IR transition at low temperatures.
Fig. 7 (a) The branching ratio β32 for the 4I11/2 to 4I13/2 transition, (b) the A coefficient and radiative lifetime of the 4I11/2 manifold, (c) the fluorescence quantum efficiency (Q.E.) of the 4I11/2 manifold, and (d) Q.E. of the 4I11/2 to 4I13/2 transition. All as a function of temperature for 0.25 at.% Er3+:Y2O3.
The Einstein A coefficient is representative of the rate that ions decay spontaneously from an excited state to a lower state in a given transition. From the derived absorption cross sections, it is possible to calculate the A coefficient for the Erbium transition originating at the 4I11/2 manifold and terminating at 4I15/2, heretofore denoted as A31. The calculation is done using the modified Fuchtbauer-Ladenburg (F-L) equation [12]:
where σabs(λ) is the absorption cross section, n is the index of refraction of Y2O3, and Z1 and Z3 are the respective partition functions of the lower and upper manifold. From the A31 coefficient and the branching ratios, it is possible to calculate the radiative lifetime of the 4I11/2 manifold using the following relations:The calculated A coefficients and radiative lifetimes, as functions of temperature, are presented in Fig. 7(b). It is shown that the calculated radiative lifetime of the 4I11/2 manifold starts at 5.4 ms at room temperature and increases to 6.2 ms as the temperature cools to 77K.With the calculated radiative lifetimes and the measured fluorescence lifetimes, it is now possible to determine the overall fluorescence quantum efficiency of the 4I11/2 manifold (η) as well as the specific quantum yield of the 2.7 µm transition (η32) using the following relationships:
andThe calculated values for η and η32 are shown, as functions of temperature, in Fig. 7(c) and 7(d), respectively. Figure 7(c) shows that the overall fluorescence quantum efficiency of 4I11/2 increases from 47% at room temperature to a maximum of 67% at ~110 K. As temperature continues to decrease, η begins to decrease. In Fig. 7(d), η32 is shown to increase consistently from a value of 5% at room temperature to approximately 11% at 77K. However, the value for η32 does appear to level off for temperatures less than 110 K.With the calculated radiative lifetime, the branching ratios, and the measured fluorescence spectra, it is possible to derive the stimulated emission cross sections for the 2.7µm and the 1.0 µm transitions using the standard F-L equation [15]:
where Iij(λ) is the fluorescence spectra of the transition from states i to j.Figure 8 shows the stimulated emission cross sections of both transitions for room temperature and 77K. The 1.0 µm spectra presented in Fig. 8(a) show that the highest intensity peak at ~980 nm experiences a 3.5x increase when cooling from room temperature to 77K, while the lines narrow by a similar factor. The 2.7µm spectra shown in Fig. 8(b) show much more dramatic results: The highest intensity peak at ~2715 nm increases 10x over the same temperature difference, while the line width only decreases by factor of 6.
Fig. 8 Stimulated emission cross section for 0.25 at.% Er3+:Y2O3 for the 2.7µm and 1.0µm transitions at 300K and 77K.
As a check of these cross section values, the McCumber method [16] was independently used to calculate the stimulated emission cross sections of the 1.0 µm transition using the reciprocity of absorption and stimulated emission between two energy levels. This reciprocity relation is written:
where ZL and ZU are the respective lower and upper partition functions, E0 is the “zero line” energy between the lowest Stark levels of the two manifolds, h is Planck’s constant, kB is the Boltzmann constant, and T is the temperature. This method requires detailed knowledge of the energy levels of the relevant manifolds which was obtained from Gruber et al [17]. The results of these calculations, using the 0.25 at.% Er3+:Y2O3 data, was an agreement within 5% of the peak cross section values presented in Fig. 8.5. Summary of recent laser results
The laser parameters collected from the experimental setup shown in Fig. 1. are summarized in Tab. 1. As is seen from the Table, the cryogenic Er:Y2O3 laser can deliver 24 W CW power at 2.74 µm. The laser emits predominantly at 2.74 um, although weak emission was also observed at 2.71 and 2.76 um. The 24% slope efficiency is slightly lower than that set by theoretical limit, which equal to ~33%. Reflectivities and curvatures of the coupling mirrors were experimentally optimized for the maximum output power. The best performance was observed with the outcoupling mirror refelectivity ~85% and the radius of curvature ~100 mm. The details on the high power Er:Y2O3 mid-IR laser will be published in a separate paper. This is to our knowledge the highest power ever achieved in a bulk solid-state laser.
Table 1. Summary of cryogenic Er3+:Y2O3 laser parameters.
6. Conclusions
Spectroscopic measurements performed on a series of ceramic Erbium-doped Y2O3 samples with differing concentrations were used to show the effect of reabsorption on the spectral shape of the 4I11/2 → 4I15/2 fluorescence. These results revealed that that samples with an Erbium concentration less than 0.5 at.% would provide the most accurate spectral measurements. Subsequent measurements of absorption, fluorescence, and fluorescence lifetimes were performed on a 0.25 at.% Er3+:Y2O3 ceramic sample, and. these results were used to calculate important parameters related to the ~2.7 µm mid-IR laser transition. Among these were the branching ratio β, the radiative lifetime τrad, the fluorescence quantum efficiency, and the stimulated emission cross sections.
It was shown that the branching ratio of the 2.7 µm transition increases from roughly 0.105 to 0.17 when cooling from room temperature to 77K. The fluorescence quantum yield of this transition was shown to increase from 5% to 11% over the same temperature change. All of this led to a 10x increase in the 2715 nm peak cross section intensity at cryogenic temperatures.
Finally, updated laser results were presented for the cryogenic Er3+:Y2O3 laser shown in Fig. 1. Among these results are 24 Watts of CW output power, which to our knowledge is the highest power ever achieved in a bulk solid-state laser.
Acknowledgments
The authors wish to thank Dr. Larry D. Merkle for helpful discussions and insights pertaining to this work.
References and links
1. T. M. Taczak and D. K. Killinger, “Development of a Tunable, Narrow-Linewidth, CW 2.066-µm Ho:YLF Laser for Remote Sensing of Atmospheric CO2 and H2O,” Appl. Opt. 37(36), 8460–8476 (1998). [CrossRef] [PubMed]
2. V. A. Serebryakov, E. V. Boiko, N. N. Petrishchev, and A. V. Yan, “Medical Applications of mid-IR lasers. Problems and Prospects,” J. Opt. Technol. 77(1), 6–17 (2010). [CrossRef]
3. S. Ishii, K. Mizutani, H. Fukuoka, T. Ishikawa, B. Philippe, H. Iwai, T. Aoki, T. Itabe, A. Sato, and K. Asai, “Coherent 2 microm differential absorption and wind lidar with conductively cooled laser and two-axis scanning device,” Appl. Opt. 49(10), 1809–1817 (2010). [CrossRef] [PubMed]
4. G. J. Kintz, R. Allen, and L. Esterowitz, “CW and pulsed 2.8µm laser emission from diode-pumped Er3+:LiYF4 at room temperature,” Appl. Phys. Lett. 50(22), 1553–1555 (1987). [CrossRef]
5. J. S. Liu, J. J. Liu, and Y. Tang, “Performance of a diode end-pumped Cr, Er: YSGG laser at 2.79um,” Laser Phys. 18(10), 1124–1127 (2008). [CrossRef]
6. Y. H. Park, H. J. Kong, Y. S. Kim, and G. U. Kim, “2.70 μm emission Er:Cr:YSGG laser with LINbO3 Pockels cell,” Laser Phys. Lett. 6(3), 198–202 (2009). [CrossRef]
7. T. Sanamyan, “Diode pumped cascade Er:Y2O3 laser,” Laser Phys. Lett. 12(125804), 1–6 (2015).
8. N. Ter-Gabrielyan, L. D. Merkle, A. Ikesue, and M. Dubinskii, “Ultralow quantum-defect eye-safe Er:Sc2O3 laser,” Opt. Lett. 33(13), 1524–1526 (2008). [CrossRef] [PubMed]
9. K. Petermann, “Oxide laser crystals doped with rare earth and transition metal ions,” in Handbook of Solid-State Lasers: Materials, Systems, and Applications (Woodhead Publishing Limited, 2013).
10. T. Sanamyan, M. Kanskar, Y Xiao, D Kedlaya, and M Dubinskii, “High power diode-pumped 2.7-µm Er3+:Y2O3 laser with nearly quantum defect-limited efficiency,” Opt. Express 19(S5), A1082–A1087 (2011). [CrossRef] [PubMed]
11. T. Sanamyan, J. Simmons, and M. Dubinskii, “Efficient cryo-cooled 2.7µm Er3+:Y2O3 ceramic laser with direct diode pumping of the upper laser level,” Laser Phys. Lett. 7(8), 569–572 (2010). [CrossRef]
12. V. Ter-Mikirtychev, Fundamentals of Fiber Lasers and Fiber Amplifiers (Springer, 2014).
13. D. K. Sarder, K. L. Nash, R. M. Yow, and J. B. Gruber, “Absorption intensities and emission cross section of intermanifold transition of Er3+ in Er3+:Y2O3 nanocrystals,” J. Appl. Phys. 101(11), 113115 (2007). [CrossRef]
14. E. E. Brown, U. Hommerich, A. Bluiett, C. Kucera, J. Ballato, and S. Trivedi, “Near-infrared and upconversion luminescence in Er:Y2O3 ceramics under 1.5 µm excitation,” J. Am. Ceram. Soc. 97(7), 2105–2110 (2014). [CrossRef]
15. S. A. Payne, L. L. Chase, L. K. Smith, L. K. Wayne, and W. F. Krupke, “Infrared cross-section measurements for crystals doped with Er3+, Tm3+, and Ho3+,” IEEE J. Quantum Electron. 28(11), 2619–2630 (1992). [CrossRef]
16. D. E. McCumber, “Einstein relations connecting broadband emission and absorption spectra,” Phys. Rev. 136(4A), A954–A957 (1964). [CrossRef]
17. J. B. Gruber, K. L. Nash, D. K. Sarder, U. V. Valiev, N. Ter-Gabrielyan, and L. D. Merkle, “Modeling the optical transitions of Er3+ in C2 and C3i sites in polycrystalline Y2O3,” J. Appl. Phys. 104, 023101 (2008). [CrossRef]
