|Home | About | Journals | Submit | Contact Us | Français|
We report the near to middle infrared luminescence and energy transfer process of Er3+/Yb3+ co-doped fluorotellurite glasses under 980, 1550 and 800 nm excitations, respectively. Using a 980 nm laser diode pump, enhanced 1.5 and 2.7 μm emissions from Er3+:I13/2→4I15/2 and I11/2→4I13/2 transitions are observed, in which Yb3+ ions can increase pumping efficiency and be used as energy transfer donors. Meanwhile, Yb3+ can also be used as an acceptor and intensive upconversion luminescence of around 1000 nm is achieved from Er3+:I11/2→4I15/2 and Yb3+: F5/2→4F7/2 transitions using 1550 nm excitation. In addition, the luminescence properties and variation trendency by 800 nm excitation is similar to that using 1550 nm excitation. The optimum Er3+ and Yb3+ ion ratio is 1:1.5 and excess Yb3+ ions decrease energy transfer efficiency under the two pumpings. These results indicate that Er3+/Yb3+ co-doped fluorotellurite glasses are potential middle- infrared laser materials and may be used to increase the efficiency of the silicon solar cells.
Over the past decades, Er3+/Yb3+ co-doped materials have attracted interest because of its usefulness for near to middle infrared emissions1,2,3,4. Erbium ion is an ideal luminescent center for 1.5 and 2.7 μm emissions, which correspond to the 4I13/2→4I15/2 and 4I11/2→4I13/2 transitions respectively5,6. The Er3+ doped fiber amplifier is one of the most important devices used in the 1.5 μm wavelength optical communication window7, and 2.7 μm emission also concerns researchers because of its possible applications in medicine, sensing, military countermeasures, and in light detection and ranging8,9,10. The absorption band of the Er3+:4I15/2→4I11/2 transition around 980 nm characterize weak ground state absorption and the sensitization of the Er3+ ions using Yb3+ ions increase the pumping efficiency3, as shown in Fig. 1(left). The Er3+/Yb3+ doped materials are used for effective energy transfer mechanisms for obtaining 1.5 and 2.7 μm emissions under 980 nm excitation, in which the Yb3+ ions are donors that transfer energy from the 2F5/2 level to the Er3+:4I11/2 level11,12. On the other hand, Yb3+ ions are acceptors when the Er3+/Yb3+ doped materials are excited using a 1550 nm Laser Diode (LD)2,13, as shown in Fig. 1(middle). The ions on the 4I15/2 ground state absorb two photons to the 4I9/2 levels. Then the ions on the 4I9/2 level decay radiatively or nonradiatively to the 4I11/2 level and the 1 and 2.7 μm emissions occur. The intensive 1000 nm upconversion luminescence converted from 1550 nm IR light in the Er3+/Yb3+ doped materials increase the efficiencies of Si solar cells because Si solar cells show highest efficiencies at 1000 nm wavelength, whereas only 60% of the visible light can be converted to electrons14,15. Meanwhile, the Er3+ ions can also be pumped directly to the 4I9/2 level under 800 nm excitation as shown in Fig. 1(right) and the luminescence properties should be similar to that under 1550 nm excitation theoretically. It is important to understand the luminous mechanism under different excitations for the Er3+/Yb3+ co-doped glasses in order to obtain more luminous information.
As host material for near to middle infrared emissions, glass attracts much research and development interest due to its ease of fabrication and its use as diode-pumped high-power solid state laser hosts, sensors, and optical amplifiers16,17,18,19. At present, attention have mainly been paid to the 2.7 μm emission of Er3+ doped fluoride glasses20 and 1 μm emission of Er3+/Yb3+ doped crystals13,21. As well known, so far most of works about 2.7 μm emission materials have been done in fluoride (ZBLAN) glasses. The glass family is the most stable one among all fluoride systems reported so far. In the past decade, Er doped and Er/Pr co-doped ZBLAN flbers have been developed for obtaining higher power output. But, the Tg of the ZBLAN is as low as 270°C which causes thermal effect. Additionally, because of the small value of ΔT, crystallization is an obstacle of fabricating high concentration ZBLAN fibers. These weaknesses limit the application of the ZBLAN in the future22. So it is important and challenging for researcher to find new mid-infrared materials. Fluorotellurite glass is a potential near to middle infrared laser material because it combines the advantages of fluoride and oxide glasses. Fluorotellurite glasses possess relatively low phonon energy among all the oxide glasses, a broad transmission window of 0.4~6 μm, and stable chemical and physical properties relative to fluoride glasses, such as easy fibering23. However, no works concern near to middle infrared emissions of Er3+/Yb3+ doped fluorotellurite glass excited under different wavelengths.
A new kind of fluorotellurite glass was prepared using AlF3- based glass modified with TeO2. Our previous work has reported the good thermal stability, low phonon energy and wide high transmittance of the glass24. In this study, near to middle infrared emissions of Er3+/Yb3+ doped glasses were measured under different excitations and the energy transfer processes between the two ions were determined. The optimum ratio of the two ions was chosen to obtain intensive 2.7 and 1 μm emissions. In addition, cross sections for the emissions and the energy transfer microparameters between the two ions were calculated.
The investigated glass has the following molar composition: 90(AlF3-YF3-CaF2-BaF2-SrF2-MgF2)–10TeO2–1ErF3-xYbF3 (x = 0, 0.5, 1, 1.5, 2, labelled as FEYx, respectively). All the samples were prepared using high-purity AlF3, YF3, CaF2, BaF2, SrF2, MgF2, TeO2, ErF3 and YbF3 powders. Well-mixed 25 g batches of the samples were placed in platinum crucibles and were melted at about 950°C for 30 min. Then the melts were poured onto a preheated copper mold and annealed in a furnace at around the glass transition temperature. The annealed samples were fabricated and polished to 20 mm × 15 mm × 1 mm dimensions for the optical property measurements.
The characteristic temperatures (temperature of glass transition Tg and temperature of the onset of the crystallization peak Tx) of the samples were determined using a NetzschSTA449/C differential scanning calorimeter at the heating rate of 10 K/min. The densities and refractive indices of the samples were measured using the Archimedes method, with distilled water as the immersion liquid and the prism minimum deviation method respectively. Furthermore, the absorption spectra were recorded using a Perkin-Elmer Lambda 900 UV/VIS/NIR spectrophotometer at the range of 300 nm to 2000 nm, and the emission spectra were measured using a Triax 320 type spectrometer (Jobin-Yvon Co., France). All the measurements were performed at room temperature.
Fig. 2 shows the absorption spectra of the Er3+- doped and Er3+/1.5Yb3+ co-doped samples at room temperature in the wavelength region of 400 nm to1600 nm. The introduction of 1.5 mol% Yb3+ greatly enhances the absorption coefficient at around 980 nm, resulting in the efficient absorption of the pump source at around 980 nm. The radiative transition parameters within the 4fn configuration of the Er3+ ions can be analyzed using the Judd-Ofelt (J-O) theory and can be accurately measured using absorption spectra25,26. The J-O parameters and radiative transition parameters (spontaneous transition probability A, branching ratio β, and calculated lifetime τ) for the Er3+:4I11/2→4I13/2 transition of the present FE and FEY1.5 samples have been calculated, and are shown in Table 1. Ω2 parameters indicate the amount of the covalent bond, and are strongly dependent on the local environment of the ion sites, whereas the Ω6 parameter is related to the overlap integrals of the 4f and 5d orbits27,28. The higher Ω2 of the codoped sample indicates a higher covalency and lower symmetry. The spectroscopic quality factor i.e. Ω4/Ω6 is an important parameter to predict the stimulated emission in a laser active host29. The spectroscopy quality factor (1.31) in the co-doped sample is much larger than those reported in fluoride glasses30, indicating that the co-doped sample is a favorable optical material. The predicted spontaneous emission probability for the 2.7 μm emission is larger in the co-doped sample, which provides a better opportunity to obtain laser actions31.
Figure 3 shows the emission spectra around 1.5 and 2.7 μm and the measured decay lifetimes of the 1.5 μm emission before and after Er3+ co-doped Yb3+ under 980 nm excitation. The intensities of the two emissions initially increase as Yb3+ ions increase, whereas the excess Yb3+ ions (>1.5 mol %) reduce the intensities. The 1.5 and 2.7 μm emissions originate from the Er3+:4I13/2→4I15/2 and 4I11/2→4I13/2 transitions, and the intensities of the two emissions are enhanced 12 and 2 times, respectively, when the ratio of the Er3+ and Yb3+ ions is 1:1.5. The Yb3+ ions positively affect the 1.5 and 2.7 μm emissions of Er3+ ions under 980 nm excitation, and the energy transfer process is from the Yb3+:2F5/2 to the Er3+:4I11/2 level. The change of the decay lifetimes with the Yb3+ contents coincides with those of the emissions. The lifetime is an important factor for potential laser materials. The full width at half maximum (FWHM)32 determines 1.5 μm laser materials. The larger bandwidth of this transition is suitable for tunable lasers that deliver relatively constant power over a wide wavelength range. The FWHM value in the EY1.5 glass in this study is about 55 nm, which is not only higher than those of silicate (34.8 nm)33 and phosphate (46.0 nm)33 but also higher than those of pure fluoride ZELAG (46 nm)34 and tellurite glasses (53 nm)35. The emission cross section is an important parameter for 2.7 μm emission which can be calculated as10,36,37
where λ is the wavelength. Arad is the spontaneous transition probability. I(λ) is the emission spectrum, n and c are the refractive index and light speed in vacuum respectively. The emission cross section of the 4I11/2→4I13/2 transition of the EY1.5 sample is calculated to be 8.3 × 10−21 cm2, which is higher than those of Er3+ doped ZBLAN glass (5.4 × 10−21 cm2)29, chalcohalide glass (6.6 × 10−21 cm2)29, fluorophosphate glass (7 × 10−21 cm2)38, and tellurite glass (6.1 × 10−21 cm2)39.
Figure 4 shows the emission spectra around 1 and 2.7 μm before and after Er3+ co-doped Yb3+ under 1550 nm excitation. The upconversion luminescence bands centered at 980 nm is a two- photon process and originate from the Er3+:4I11/2→4I15/2 and Yb3+:4F5/2→4F7/2 transitions. After introducing Yb3+, the line shapes of the emission from the co-doped samples significantly change and are similar to those recorded from other materials containing Tb3+/Yb3+ and Pr3+/Yb3+ 40,41, which indicates that the emission is probably due to the transition in the Yb3+ ions and the energy transfer process from the Er3+:4I11/2 to the Yb3+:2F5/2 level. The intensity of the emission is highest when the Er3+ and Yb3+ ratio is 1:1.15. The energy transfer process cannot be efficiently performed with excess Yb3+ ion content the content. The obvious 2.7 μm emission is observed in the Er3+ singly doped sample and it is hardly to be obtained in the co-doped samples which can be explained by the energy transfer from the Er3+:4I11/2 level to the Yb3+:2F5/2 level.
Figure 5 shows the emission spectra around 1 and 2.7 μm before and after Er3+ co-doped Yb3+ under 800 nm excitation. The similar phenomenon has been observed as that under 1550 nm excitation. The 1 μm is enhanced significantly in the co-doped samples and the 2.7 μm emission is decreased with the increasing Yb3+ ions when the Yb3+ content is below 1.5 mol %, which demonstrate the Yb3+ ions accept the energy from the Er3+ ions and the 1 μm emission mainly comes from the Yb3+:4F5/2→4F7/2 transition. Figure 5 also shows the energy transfer process can proceed efficiently when the ratio of the Er3+ and Yb3+ ions is 1:1.5.
As discussed above, the Yb3+:2F5/2 level can transfer energy to the Er3+:4I11/2 and the backward process can also occur. If energy of the emission transition of one RE3+ ion (called the donor) is equal or close to the energy of the absorption transition of the other RE3+ ion (called the acceptor). Energy transfer between rare earth ions can occur according to the Forster and Dexter theory42,43,44. The absorption cross section of the Er3+:4I11/2→4I15/2 and Yb3+:4F7/2→4F5/2 transitions can be deduced by the Beer-Lambert equation43
where is the absorptivity from absorption spectrum, l is the thickness of the glass and N is the ion density.
The emission cross section can be obtained by using the McCumber equation3:
where h is Planck's constant, KB is the Boltzmann constant, T is the temperature, Ezl is the ground state manifold and the lowest stark level of the upper manifolds and Zu and Zl are partition functions of the lower and upper manifolds.
Fig. 6 shows the absorption and emission cross sections of the Yb3+ and Er3+ ions. Figure (a) describes the energy transfer process from Yb3+ to Er3+, which corresponds to the results under 980 nm excitation, and (b) describes the reverse process which corresponds to the results under 1550 and 800 nm excitations. The overlap between σa and σe is quite large, therefore, efficient energy transfer can be expected between the two ions.
Based on the obtained absorption cross section of the donor and the emission cross section of the acceptor, the probability rate of the energy transfer between the donor and the acceptor can be described as
where is the matrix element of the Hamiltonian perturbation between the initial and final states in the energy transfer process, and is the overlap integral between the m-phonon emission line shape of the donor ions (D) and the k-phonon emission line shape of the donor ions (A). In the case of weak electron-phonon coupling, can be approximated as
where SDA (0, 0, E) represents the overlap integral between the zero-phonon line shape of the donor emission ions and the absorption of the acceptor ions, and S0D, and S0A are the Huang-Rhys factors of the donor and acceptor ions, respectively. The probability rate of the energy transfer can be obtained using the following direct transfer equation:
where CD-A is the energy transfer coefficient, and R is the distance of separation between the donor and acceptor, and the critical radius of the interaction can be obtained by the equation , and τD is the intracenter lifetime of the excited level of the donor. The expression for direct transfer (D-A) is expressed as:
The microparameters of energy transfer from Yb3+:2F5/2 to the Er3+:4I11/2 and the reverse process are calculated using Eqs (4)–(7). The values are 2.06 × 10−39 and 2.12 × 10−39 cm6/s, respectively, and they both are independent of phonon in the quasiresonant process. These results show that the high energy transfer process efficiency between the two ions and the direction of the process are dependent on the excitations. The Er3+/Yb3+ co-doped fluorotellurite glasses can be used to obtain 1.5 and 2.7 μm emissions under 980 nm excitation and 1000 nm upconversion luminescence under 1550 nm excitation.
In conclusion, Er3+/Yb3+ co-doped fluorotellurite glasses are prepared, and their near to middle infrared emissions under different excitations were investigated. The 1.5 and 2.7 μm emissions are effectively enhanced by the presence of Yb3+ ions under 980 nm excitation. The FWHM of the 1.5 μm emission is as high as 55 nm and the emission cross section of the 2.7 μm emission reaches 8.3 × 10−21 cm2. Under 1550 nm excitation, the Yb3+ ions accept the energy from the Er3+ ions, and the upconversion luminescence centered at λ = 980 nm from the co-doped samples is enhanced. The optimum Er3+ and Yb3+ ion ratio in this system is 1:1.5, under different excitations. The energy transfer microparameters from the Er3+ to Yb3+ and the reverse process are calculated to be 2.06 × 10−39 and 2.12 × 10−39 cm6/s, respectively. Therefore, the 1.5 μm emission is useful optical communication window, 2.7 μm has possible applications in medicine, sensing, and military countermeasures, and 980 nm upconversion luminescence corresponds to the most efficient absorption wavelength of Si solar cells and can be used to increase their efficiency.
This work is financially supported by National Natural Science Foundation of China (No. 51172252, and 51272262).
The authors declare no competing financial interests.
Author Contributions F.H. wrote the main manuscript text and coauthor X.L., Y.M. and S.K. checked up. D.C. and L.H. are responsible for the experiment. All authors reviewed the manuscript.