We report for the first time the characterization of glass-ceramics for optical refrigeration. Ytterbium-doped nanocrystallites were grown in an oxyfluoride glass matrix of composition 2YbF3:30SiO2-15Al2O3-25CdF2-22PbF2-4YF3, forming bulk glass-ceramics at three different crystalisation levels. The samples are compared with a corresponding uncrystalised (glass) sample, as well as a Yb:YAG sample which has presented optical cooling. The measured X-ray diffraction spectra, and thermal capacities of the samples are reported. We also report for the first time the use of Yb:YAG as a reference for absolute photometric quantum efficiency measurement, and use the same setup to characterize the glass and glass-ceramic samples. The cooling figure-of-merit was measured by optical calorimetry using a fiber Bragg grating and found to depend on the level of crystallization of the sample, and that samples with nanocrystallites result in higher quantum efficiency and lower background absorption than the pure-glass sample. In addition to laser-induced cooling, the glass-ceramics have the potential to serve as a reference for quantum efficiency measurements.
© 2015 Optical Society of America
Doppler cooling involves translational cooling of noninteracting atoms , while in anti-Stokes cooling the high-energy emitted photon is used to remove thermal energy from the ion’s surrounding. The possibility of using anti-Stokes fluorescence for optical cooling was predicted by Pringsheim in 1929 . The idea of optical cooling seemed to violate the second law of thermodynamics, but this question was settled with the mathematical formulation of the problem by Landau . Optical refrigeration of solids can achieve cooling efficiencies of more than 104 times larger than those observed in laser Doppler cooling of gases , thus allowing the use of a laser to refrigerate relatively massive loads to cryogenic temperatures [5,6], and the applications are quite different from those of Doppler cooling.
Optical cooling combines some of the advantages of thermoelectric coolers, e.g. miniaturization and potential volume production, with those of mechanical coolers, e.g. low temperature limit . In addition, an optical cooling scheme can benefit cooling systems that require electromagnetic radiation and mechanical insulation, for lightweight cryogenics  and in situ cooling . Rare earth doped fluoride crystals and glasses have been the workhorse materials for optical refrigeration of solids . They exhibit low phonon energy, desirable for reducing non-radiative decays or quenching . The fluoride glass ZBLAN was the first solid-state material to be optically cooled , but it has a higher phonon energy than other hosts used for cooling and it is a challenging material to prepare . For instance ZBLAN has a maximum phonon energy of 580 cm−1 , while in PbF2 crystals it is only 250 cm−1 . The most often used optical cooling material to date is Yb:YLiF4, which has been optically cooled to cryogenic temperatures . High-purity single crystals are hard to mass-produce, and in the case of fluorides, are also hard to handle and polish , since they can be brittle and prone to chemical attack.
Nanocrystals of low phonon energy such as fluorides have gained attention recently mainly for their strong up-conversion processes due to weak quenching and strong absorption. These characteristics are desirable and useful for displays [15,16], lasers, biomarkers , bioactuators , photovoltaics , and temperature sensors [20–22], to cite a few. A common route to use these properties in a robust material is to create a composite which shares the desirable optical properties of the nanocrystals with the mechanical and chemical properties of a glass host. In the case of upconversion applications, usually the crystals are co-doped with multiple types of ions for enhancing the absorption and enabling different emissions . For laser cooling applications, the same properties of the host are desirable, but a single kind of dopant with a large bandgap is usually a safer way to avoid nonradiative decays, thus increasing the cooling efficiency , although multilevel cooling or the use of dopants with lower bandgaps than in Ytterbium are also useful candidates [25,26].
Fiber Bragg gratings have proved to be an ideal sensor for contact temperature measurement in laser cooling experiments, being able to record temperature during continuous, or pumping with low powers . Here we use this method to assess the potential of optical cooling of glass-ceramic samples by measuring the temperature changes as a function of the pump wavelength. As an additional method to measure quantum efficiency, we used the integrating sphere method , to characterize the samples, and for the first time to our knowledge, the setup was assessed using a sample of Yb:YAG that had already demonstrated optical cooling.
Our results show that our glass-ceramics have a higher cooling figure-of-merit than their glassy counterparts. With further purification, glass-ceramics are expected to show cooling and can pave a new way towards mass applications of optical refrigeration.
The cooling power, or the power which is extracted from a sample by optical means – excluding blackbody radiation – is given by the difference between the optical power which leaves the sample and the optical power that enters the sample, and can be expressed by (1) , without considering radiation trapping or reabsorption, or including these in the quantum efficiency.30]. It can be practical to define a theoretical cooling efficiency of a material at a given pump wavelength as the ability of conversion of the absorbed power into heat lift or cooling power :
For simplicity, considering a radiatively grey sample at temperature Ts enclosed by a blackbody at temperature Tr, one can find that the sample will exchange heat with the environment at a rate Pload given by the Eq. (4) .Eq. (4) can be reduced, without loss of generality, to the form (5).
3. Material preparation and structural characterization
Oxyfluoride glass with chemical composition 30SiO2-15Al2O3-27CdF2-22PbF2-4YF3-2YbF3 was prepared by conventional melt-quenching technique. High purity (Sigma-Aldrich, 99.99%) precursors were well mixed in an agate mortar to obtain homogeneous powder, and were transferred to a covered platinum (Pt-100%) crucible. The crucible was fired at 1100 °C for 2 h in a muffle furnace. The resulting molten glass was poured into a stainless steel mold that was pre-heated at the glass transition temperature, where the glass was annealed at the same temperature for 5 h to remove internal stresses and strains. The solid samples were cut and polished optically to dimensions 10 × 2 × 2 mm3 for spectroscopic and cooling measurements. Some of the samples were not polished, but used for thermal and structural characterization.
To obtain the glass transition, onset crystallization and peak crystallization temperatures, we used a Netzsch DSC 404F3 Pegasus® to perform differential scanning calorimetry (DSC) on the unpolished sample. The DSC trace of the as-made 2 mol% Yb3+-doped oxyfluoride glass is shown in Fig. 1.
From the trace shown in Fig. 1, one can obtain the glass transition temperature Tg = 408 °C, the onset crystallization temperature Tx = 433 °C and the peak crystallization temperature Tp = 457 °C. The glass samples were annealed at different temperatures and timescales to transform them into glass-ceramics (GC’s). The whole synthesis was performed at ambient air at atmospheric pressure. The obtained GC’s are ultra-transparent in nature and their details and labels are shown in Table 1. After fine calibration with a sapphire reference as well as fused silica, the same setup was used to measure the thermal capacity of glass sample, the GC3, and a glass-ceramic sample annealed for 25 h at 470 °C, which is used as reference for the GC1 and GC2.
In order to estimate the effect of the annealing time and temperatures on the crystallization, we performed X-ray diffraction (XRD) measurements on samples of the same preparation as the YbG, GC1 and GC2, using a X-ray Bruker D8 Advance (CuKα12 radiation) equipped with a Vantec-1 linear detector. The resulting XRD profile is shown in Fig. 2.
The glass-ceramics are free from any secondary phase crystals and uniformly dispersed β-PbF2 nanocrystals were grown after appropriate thermal treatment. The XRD profile of glass contains two broad curves typical of amorphous structure whereas the GCs contain several sharp peaks related to the diffraction pattern of the crystalline structure, which can be identified as cubic β-PbF2 nanocrystalline phase. The amount of crystalline phase improves with increasing the heat treatment temperature. The crystallite size of the nanocrystals was evaluated by using the Scherrer’s formula from the XRD peak widths and was found to be less than 20 nm.
4. Quantum efficiency using integrating sphere
The measurement of quantum efficiency is extensively performed for the characterization of fluorophores for biomedical applications, LEDs for illumination, etc. It is however an uncommon measurement in laser science as a whole field. Due to the stringent requirements of optical refrigeration, the knowledge of the quantum efficiency of the materials being studied is of fundamental importance. The external quantum efficiency is defined as33]. When the sample is solid, the problem is more complicated since it is not always possible to have a reference with same optical geometry, and a full-spherical measurement may be needed such as by an integrating sphere. Since the geometry and dilution of fluorophores – or active ions – are limited, reabsorption effects are hard to exclude.
The method introduced by , which we use here, consists in measuring the relative spectral photon number density in the detector in three situations: A, when the pump beam is directed towards the empty Ulbricht integrating sphere (ISp); B, when the sample is inside the ISp and the beam is directed onto the ISp wall; C, when the sample is inside the ISp and the beam is directed onto the sample. The setup which we utilized is represented in Fig. 3. The output from a Ti:Sapphire is kept at a constant p-polarization by a Glan-Thompson polarizer, then ~1% is tapped by a glass plate, and the reflection is monitored by a Keithley 6487 Picoammeter. The transmitted beam with ~200 mW of power is focalized at the integrating sphere entrance port and directed to the center of the sphere. The diffused light from the sphere walls is collected by a multimode fiber of 200 μm diameter and directed to an Ando AQ6317B optical spectrum analyser. The data is collected and processed by a computer, which measures 50 spectra, while normalizing them to the tapped optical power, between the measurements of two spectra..
The detected numbers of photons are spectrally separated in two regions: laser and photoluminescence, and are labelled respectively by Li and Pi, with i being A, B or C, depending on the setup configuration as described above. If the original measurand is spectral power density, it should be multiplied by the corresponding wavelength in order to have the proportional spectral number of photons density. The quantum efficiency can be calculated using (9).34], while it is generally assumed to be 0.0050. In order to avoid such uncertainties, we used a sample of 3%:Yb:YAG crystal, which had its ηext measured to be 0.9914 by a calorimetric method . A similar sample had ηext = 0.988 measured by photothermal deflection . These values are very reliable, due to the strict conditions which the sample needed to meet in order to achieve optical refrigeration.
5. Optical calorimetry
The experimental setup to measure the cooling figure of merit is an optical calorimeter where the sample temperature dynamics under pumping is measured using a fiber Bragg grating. The sample is held by four 1 cm long, 125 μm diameter silica fibers attached to 1 mm diameter steel posts, inside a black-painted aluminum chamber with 2 silica windows. The chamber is purposely non-hermetic and without active thermal control thus kept at atmospheric pressure and around room temperature. For pumping, we used a tunable CW Ti:Sapphire laser, coupled to a single mode fiber (HI 1060 Flex), and recollimated into the sample. A thermocouple TC is used to measure the exact chamber temperature. A fiber Bragg grating (FBG) is placed on the top of the sample, to measure its temperature using a Bragg wavelength interrogator and a single frequency laser. The interrogator consists in a tunable laser used as probe, while two power meters monitor the transmitted power from the FBG or the reflected power from the optical circulator connected to the probe and the FBG input. The interrogator is described in detail in . The complete experimental setup is depicted in Fig. 4.
The temperature dynamics are recorded over time during pumping and when the shutter closes the beam. During the period in which the pump is shut-off, the Ti:Sapphire laser wavelength was manually changed using a tuning micrometer. The steady-state temperature ΔT∞ is deduced from an exponential fit to the temperature data.
The samples exhibited heating rather than cooling for the pump wavelengths used here, and their heating spectra is shown in Fig. 5. The best fit of (6) is shown in the figure as the green solid curve, while the curves corresponding to 95% confidence interval of the fitted ηc and αb are shown in dashed curves. The absorption coefficient is also plotted as a function of wavelength for a reference.
From the data in Fig. 5, it can be observed that the glass-ceramics presented higher quantum efficiency and smaller background absorption than their glass precussor. In the long wavelength tail of the spectra, where the resonant absorption is weak, YbGlass showed the highest temperature, which is due to background absorption. At the same time, a temperature rise of 40 K/W is observed in the glass when pumped at 1020 nm, a level which is not observed in the YbGC1 even at the 1000 nm wavelength, where the resonant absorption is much higher. The same is observed in YbGC2 and to a lesser extent in the YbGC3. That indicates a better ratio of anti-Stokes emission to non-radiative decay in the GCs when compared to the glass. Table 2 summarizes the findings and fitted data. This impact on the quantum efficiency was expected, since a fraction of the ions are within a lower phonon-energy host than the ions hosted in a glassy matrix. An explanation for the decrease in the background absorption is less trivial. In a first instance, the annealing may relief some stresses or inhomogeneities which are partially responsible for background absorption. The material structure data at the present is insufficient to determine the exact mechanism responsible for the reduction in αb. With further annealing, the size of the crystallites increases, and so does the scattering increase, leading to an indirect increase in the background absorption.
The same fitted exponential used to calculate the equilibrium temperature can be used to obtain the thermal equilibration time τh which together with the sample’s thermal capacity, can be used to deduce the sample-environment thermal coupling coefficient κ. This allows us to better evaluate the material rather than the physical sample, by calculating the materials cooling figure of merit (CFOM), shown in Fig. 6.
The cooling figures-of-merit shown in Fig. 6 reproduces the same qualitative behavior as seen in Fig. 5, but without the ambiguity that the samples’ dimensions and thermal coupling with the environment could influence if they were dissimilar. Evidently, the quantitative behavior, thus the cooling parameters are the same as before, since the model already takes the thermal coupling into account. From the present results, it is obvious that the samples need improvement before showing a positive cooling figure-of-merit. The red curve on Fig. 6 shows what would be the CFOM if the quantum efficiencies were higher by 0.03 and the background absorption lower by a factor 50, which would lead GC1 to cool by 1.75 K/W. If ηc were as high as in the Yb:YAG, GC1 could cool by −3.1 K/W, which corresponds to a final temperature difference of −13.0 K, for the same 4.2 W of pump power which drove the Yb:YAG sample 8.8 K below the chamber temperature. While the glass sample would still heat, it is clear that the YbGC1 and YbGC2 would cool, showing that the low-temperature glass-ceramic performs better than the glass. Given the present exploratory preparation conditions, these factors are feasible with further purification of the precursors. For instance purification methods developed for cooling have achieved better than 1000-fold reduction in transition metal contamination in ZBLAI glass , and intense research has achieved similar results in crystal purification as well . On the other hand, glass-ceramics are much easier to manufacture on a large scale than are single crystals. The optimal cooling wavelength for YbGC1 is 1025 nm, and for ηext < 0.975 no cooling is possible at this pump wavelength, independently of the value of the background absorption. Due to this specifity, it is worth mentioning that, in addition to optical refrigeration, these samples are strong candidates for quantum efficiency references at the near-infrared region, if indeed optical refrigeration becomes a reality with them.
We have fabricated ytterbium-doped transparent glass and glass ceramics and measured their thermal and spectroscopic properties. This work presents for the first time, the measured cooling figure-of-merits of these samples. The calorimetric results on the quantum efficiency of the samples are in line with the results obtained using an integrating sphere, showing an above 0.9 quantum efficiency for all the samples measured. It has also been shown for the first time, the use of an optically-cooled sample as an accurate reference for quantum efficiency measurements. We estimate that the use of a glass-ceramic could lead to an accuracy in the measurement of QE to better than 2%. As expected, nonradiative decays are considerably reduced in the glass-ceramics when compared to those seen in glass. Hence, rare earth doped glass-ceramics offer a promising path towards samples with higher efficiency for use in optical refrigeration and metrology, with a much more scalable production method than crystals. Also shown here is, that further precursor purification will ultimately lead to samples which may exhibit laser induced cooling.
RK acknowledges support from the Natural Sciences and Engineering and Research Council (NSERC) of Canada’s Strategic grants program, NSERC’s Discovery Grants program, Canada Council for the Arts’ Killam Research Fellowships program, and the Government of Canada’s, Canada Research Chairs program.
References and links
1. C. E. Mungan and T. R. Gosnell, “Laser cooling of solids,” Adv. At. Mol. Opt. Phys. 40, 161–228 (1999). [CrossRef]
2. P. Pringsheim, “Zwei bemerkungen über den unterschied von lumineszenz- und temperaturstrahlung,” Zeitschrift Für Physik A Hadrons and Nuclei. 57, 739–746 (1929).
3. L. Landau, “On the thermodynamics of photoluminescence,” J. Phys. USSR 10(1946).
4. R. I. Epstein, M. I. Buchwald, B. C. Edwards, T. R. Gosnell, and C. E. Mungan, “Observation of laser-induced fluorescent cooling of a solid,” Nature 377(6549), 500–503 (1995). [CrossRef]
5. D. V. Seletskiy, M. P. Hehlen, R. I. Epstein, and M. Sheik-Bahae, “Cryogenic optical refrigeration,” Adv. Opt. Photon. 4(1), 78–107 (2012). [CrossRef]
6. S. Melgaard, D. Seletskiy, V. Polyak, Y. Asmerom, and M. Sheik-Bahae, “Identification of parasitic losses in Yb:YLF and prospects for optical refrigeration down to 80K,” Opt. Express 22(7), 7756–7764 (2014). [CrossRef] [PubMed]
7. G. Mills and A. Mord, “Performance modeling of optical refrigerators,” Cryogenics 46(2-3), 176–182 (2006). [CrossRef]
8. G. Nemova and R. Kashyap, “Fiber amplifier with integrated optical cooler,” J. Opt. Soc. Am. B 26(12), 2237–2241 (2009). [CrossRef]
9. M. P. Hehlen, M. Sheik-Bahae, R. I. Epstein, S. D. Melgaard, and D. V. Seletskiy, “Materials for optical cryocoolers,” J. Mater. Chem. C 1(45), 7471–7478 (2013). [CrossRef]
11. X. Zhu and N. Peyghambarian, “High-power zblan glass fiber lasers: Review and prospect,” Adv. Optoelectron. 2010, 1–23 (2010). [CrossRef]
12. S. Aasland, M.-A. Einarsrud, T. Grande, and P. F. McMillan, “Spectroscopic investigations of fluorozirconate glasses in the ternary systems ZrF4-BaF2-AF (a= Na, Li),” J. Phys. Chem. 100(13), 5457–5463 (1996). [CrossRef]
13. V. K. Tikhomirov, G. Adamo, A. E. Nikolaenko, V. D. Rodriguez, P. Gredin, M. Mortier, N. I. Zheludev, and V. V. Moshchalkov, “Cathodo- and photoluminescence in Yb(3+)-Er(3+) co-doped PbF(2) nanoparticles,” Opt. Express 18(9), 8836–8846 (2010). [CrossRef] [PubMed]
14. B. Zhong, Y. Jia, L. Chen, Y. Deng, Y. Chen, and J. Yin, “Cavity-enhanced laser cooling for Yb3+-doped fluoride crystal using a low-power diode laser,” J. Opt. Soc. Am. B 31(9), 2116–2120 (2014). [CrossRef]
15. F. Vetrone and J. A. Capobianco, “Lanthanide-doped fluoride nanoparticles: Luminescence, upconversion, and biological applications,” Int. J. Nanotechnol. 5(9/10/11/12), 1306–1339 (2008). [CrossRef]
16. P. Babu, K. H. Jang, ChS. Rao, L. Shi, C. K. Jayasankar, V. Lavín, and H. J. Seo, “White light generation in Dy(3+)-doped oxyfluoride glass and transparent glass-ceramics containing CaF2 nanocrystals,” Opt. Express 19(3), 1836–1841 (2011). [CrossRef] [PubMed]
17. M. Wang, G. Abbineni, A. Clevenger, C. Mao, and S. Xu, “Upconversion nanoparticles: synthesis, surface modification and biological applications,” Nanomedicine (Lond. Print) 7(6), 710–729 (2011). [CrossRef] [PubMed]
21. F. Vetrone, R. Naccache, A. Zamarrón, A. Juarranz de la Fuente, F. Sanz-Rodríguez, L. Martinez Maestro, E. Martín Rodriguez, D. Jaque, J. García Solé, and J. A. Capobianco, “Temperature sensing using fluorescent nanothermometers,” ACS Nano 4(6), 3254–3258 (2010). [CrossRef] [PubMed]
22. W. Xu, X. Gao, L. Zheng, Z. Zhang, and W. Cao, “Short-wavelength upconversion emissions in Ho3+/Yb3+ codoped glass ceramic and the optical thermometry behavior,” Opt. Express 20(16), 18127–18137 (2012). [CrossRef] [PubMed]
23. F. Vetrone, J.-C. Boyer, J. A. Capobianco, A. Speghini, and M. Bettinelli, “Significance of yb3+ concentration on the upconversion mechanisms in codoped Y2O3:Er3+,Yb3+ nanocrystals,” J. Appl. Phys. 96(1), 661–667 (2004). [CrossRef]
25. S. R. Bowman, J. Ganem, and C. G. Brown, “Optical cooling in multi-level systems,” Proc. SPIE 9000, 90000G (2014). [CrossRef]
26. G. Nemova, E. Soares de Lima Filho, S. Loranger, and R. Kashyap, “Laser cooling with nanoparticles,” in SPIE Photonics North, (Montreal, Canada, 2012).
28. J. C. de Mello, H. F. Wittmann, and R. H. Friend, “An improved experimental determination of external photoluminescence quantum efficiency,” Adv. Mater. 9(3), 230–232 (1997). [CrossRef]
29. M. Sheik-Bahae and R. I. Epstein, “Optical refrigeration,” Nat. Photonics 2(12), 693–699 (2007). [CrossRef]
30. R. Kashyap and G. Nemova, “Laser induced cooling of solids,” Phys. Status Solidi 8(1), 144–150 (2011). [CrossRef]
31. M. Sheik-Bahae and R. I. Epstein, “Laser cooling of solids,” Laser Photon. Rev. 3(1-2), 67–84 (2009). [CrossRef]
32. E. S. L. Filho, M. Gagné, G. Nemova, R. Kashyap, M. Saad, and S. Bowman, “Sensing of laser cooling with optical fibres,” in 7th International Workshop on Fibre Optics and Passive Components, (Montréal, 2011). [CrossRef]
33. C. Würth, J. Pauli, C. Lochmann, M. Spieles, and U. Resch-Genger, “Integrating sphere setup for the traceable measurement of absolute photoluminescence quantum yields in the near infrared,” Anal. Chem. 84(3), 1345–1352 (2012). [CrossRef] [PubMed]
34. C. Würth, D. Geißler, T. Behnke, M. Kaiser, and U. Resch-Genger, “Critical review of the determination of photoluminescence quantum yields of luminescent reporters,” Anal. Bioanal. Chem. 407(1), 59–78 (2015). [CrossRef] [PubMed]
35. R. I. Epstein, J. J. Brown, B. C. Edwards, and A. Gibbs, “Measurements of optical refrigeration in ytterbium-doped crystals,” J. Appl. Phys. 90(9), 4815–4819 (2001). [CrossRef]
36. E. Soares de Lima Filho, G. Nemova, S. Loranger, and R. Kashyap, “Direct measurement of laser cooling of Yb:YAG crystal at atmospheric pressure using a fiber bragg grating,” Proc. SPIE 9000, 90000I (2014). [CrossRef]
37. W. M. Patterson, P. C. Stark, T. M. Yoshida, M. Sheik-Bahae, and M. P. Hehlen, “Preparation and characterization of high-purity metal fluorides for photonic applications,” J. Am. Ceram. Soc. 94(9), 2896–2901 (2011). [CrossRef]