Open Access
Add to BibSonomyAbstract
New zirconia-germano-alumino silicate, nano-particles based, Ytterbium doped fibers are obtained through the conventional modified chemical vapour deposition and solution doping techniques. The start fiber preforms are characterized by means of electron micro probe, energy dispersive x-ray, and electron diffraction analyses, revealing the presence of phase-separated nano-sized Ytterbium-rich areas in the core, while the final fibers are inspected in the sense of spectroscopy and laser properties.
©2011 Optical Society of America
1. Introduction
Considerable work has been carried out on the incorporation of rare-earth (RE) oxide nano-particles into different glass hosts. A number of techniques developed to date, such as co-sputtering [1], pyrolysis [2], ion implantation [3], laser ablation [4], sol-gel [5,6], and direct nano-particles deposition [7], brought the idea in practice. Although all the mentioned processes are related to the outside vapour deposition, except for the sol-gel technique that involves longer preform fabricating time, the solution doping (SD) technique [8] in the modified chemical vapour deposition (MCVD) process is a commoner way to incorporate RE ions into the core area of a silica fiber preform. The incorporation of RE ions into suitable nano-host, which are to be dispersed within silica rich matrix of a preform through this method, is challenging compared to the fabrication of such type bulk materials by normal crucible melting. As examples, the synthesis of Er2O3 doped nano-particles [9] and Ytterbium (Yb) doped Y2O3 nano-particles [10] into silica glass and fibers by applying the basic principles of phase separation phenomena, has been reported.
In the present work, we report, for the first time to the best of our knowledge, the incorporation of Yb (in the form of Yb2O3) into zirconia-germano-alumino (ZGA) silicate phase-separated nano-particles in the core area of fiber preforms. The preforms are inspected by means of electron probe micro-analysis (EPMA) and energy dispersive x-ray (EDX) and electron diffraction analyses, which reveal the existence of phase separation in the core glass, with nano-scaled particles rich in Yb being formed. The single-mode fibers drawn from the preforms are characterized from the points of view of spectroscopic properties and lasing potential. Notice that only a couple of works dealing with nano-structured RE doped silica-zirconia based optical fibers is known (see Ref [11]. and references therein); our present paper can serve for further insight to the matter.
The choice of such class of fibers having Yb2O3 doped phase-separated ZGA-rich nano-particles dispersed into silica core glass was to get good lasing efficiency provided by none or insignificant clustering effects among Yb dopants. ZrO2 is known for its outstanding chemical and physical properties like superior hardness, chemical stability, and thermo-mechanical resistance; due to its optical transparency, high refractive index, and photochemical stability, ZrO2 is an excellent candidate for photonics applications [12,13]. A major quenching process in RE-doped materials is multiphonon relaxation. Zirconium oxide possesses of a stretching vibration at about 470 cm−1, which is very low compared with that of Al2O3 (870 cm−1) and SiO2 (1100 cm−1) [14]. The introduction of a codopant cation such as Zr4+ allows one to avoid formation of Er3+ or Yb3+ clusters in silica and consequently to exacerbate the luminescence [15]. Both Aluminum and Zirconium ions surround the Yb ions to form a solvent shell, thereby adjusting the charge balance and improving the solubility of Yb ions in the host. Moreover, in silica glass, the Yb3+ ion acts as a network modifier, of which the positive charge is not fully compensated by the non-bridging oxygens of silica network. This would cause the clustering of Yb3+ ions, resulting in the concentration quenching. In the oxide glasses, the ionic refractivity of non-bridging oxygens is larger than the ionic refractivity of bridging oxygens [16]. So, the larger ion Zr4+ could cause an overall change in symmetry of the electron cloud surrounding oxygens. When the intermediate Al2O3 in the glasses is replaced by the modifier ZrO2, the number of non-bridging oxygen is expected to increase. The Al3+ and Zr4+ ions also act as network modifiers, which makes the silica network structure more open. Therefore, this open structure is favorable for dispersing Yb3+ ions in the glass structure, which ought to result in reducing the clustering effects that is believed to be responsible for photodarkening (PD) of Yb-doped fiber lasers.
Hence, much attention is currently paid to finding out the host materials with low phonon energy in order to improve lasing efficiency and suppress or mitigate the concentration quenching and PD in heavily Yb-doped optical fibers. In our earlier work [10], we studied yttrium-alumino-silicate glass host but in that case we were able to incorporate maximum 0.60 mol% doping levels of Y2O3 along with maximum 0.13 mol% doping levels of Yb2O3 (in excess of these values opaqueness of the glass starts). In the present work, we were successful in incorporating 1.3 mol% of ZrO2 along with 1.2 mol% of Yb2O3 without any loosing of transparency of the core glass.
2. Fiber preforms fabrication
1. Doping of ZGA silicate glass with Yb2O3 oxide was done through the SD process. Small amounts of Y2O3 and P2O5 were added into the core glass where both dopants served as nucleating agents in order to increase the phase separation with generation of Yb2O3 and ZGA rich phase-separated nano-particles. The glass formers incorporated by vapor phase deposition involved SiO2, GeO2, and P2O5 along with the glass modifiers Al2O3, ZrO2, Y2O3, and Yb2O3 incorporated by the SD technique.
A single porous layer of germano-phosphorous silica was deposited within inner surface of silica tube of OD:20 mm and ID:17 mm at optimum deposition temperature of 1375 ± 25°C with flow of a mixture of SiCl4, GeCl4, POCl3, O2, and He followed by pre-sintering for three passes with gradually increasing the temperature from 1300 to 1450°C. Soaking of the porous soot layer was done into a solution using an alcoholic-water (1:5) mixture of suitable strength of YbCl3.6H2O, AlCl3.6H2O, YCl3.6H2O, and ZrOCl2.8H2O for one hour. Such multiple pre-sintering passes are undertaken for making good adhesion of the deposited porous layer with inner silica surface to prevent any disturbances during the solution soaking period, using suitable strength of the dopants precursors or afterward. The highly porous layer exhibits a greater tendency to fracture during immersion in a high-viscous (10-15 cP) mixture of dopant precursors with water or alcohol. Such type of damage might be caused by internal stresses developed within the pores resulted from hydrogen bonding of the solvent to silica surface if porous layers are not deposited properly when maintaining good adhesion with the inner silica surface. After drain-out of the solution, the layer was dried with flow of N2 gas at room temperature for 30 minutes. After that, the soaked layer containing Yb and Al salts was dried thermally by heating at around 1100-1200°C with flow of O2 at the rate of 500 cc/min along with flow of He at the rate of 75-100 cc/min for oxidation of the Yb, Y, and Al salts.
Dehydration of the core layer was carried out within a temperature range of 800-1100°C in the presence of Cl2 and O2 where ratio of Cl2:O2 ranges from 1.5:1 to 3:1. Sintering of the core layer was done with flow of mixture of 80% O2 and 20% He in a temperature range of 1300-1900°C by a step-wise increment by 100°C. After complete sintering, the tube was collapsed with flow of mixture of 90-95% O2 and 10-5% He at a temperature between 2000 and 2300°C to obtain a preform. The tube was collapsed in 3 to 4 passes of the burner at temperatures above 2000°C. The burner traverse speed was gradually decreased from 4 to 2 cm/min in subsequent passes. Weak flow of GeCl4 was maintained during collapsing to neutralize evaporation of GeO2 from the core at higher temperature and to reduce the extent of dip at the center. Deposition of porous core layer, drying of soaked layer, sintering, and collapsing were done under highly oxidizing atmosphere to reduce the formation of defects associated with aluminum oxygen hole center (AlOHC) and Yb2+ ions. The procedure allowed us to reach the optimum parameters, targeted on final fibers with numerical aperture (NA) around 0.20–0.25.
At a very high temperature, above 2350°C, ZrO2 has a cubic structure. At intermediate temperatures, between 1170 and 2350°C it has a tetragonal structure. At low temperatures, around 1170°C, the material transforms to the monoclinic structure. The transformation from tetragonal to monoclinic structure is very rapid and is accompanied by a 3 to 5 percent volume increase that causes extensive cracking in the material. Such type of cracking within the doped core was observed after fabrication of some preforms. This behaviour destroys the mechanical properties of the fabricated components during cooling. However several oxides that dissolve in the zirconia crystal structure can slow down or eliminate these crystal structure changes. Commonly used effective additives are MgO, CaO, and Y2O3 [17]. In the present work, to prevent such type of cracking, we used minor amounts of Y2O3.
We have made two preforms (SM-1 and SM-2), having the same contents of GeO2, P2O5, Al2O3, and Y2O3 while different contents of ZrO2 and Yb2O3 to study the effect of Zr/Yb ratio on nano-structuration and the resultant fiber spectroscopic properties. For making similar doping levels of GeO2 and P2O5, we applied the deposition program with the same flow of SiCl4, GeCl4, and POCl3. For incorporating the same contents of Al2O3 and Y2O3 using SD, we deposited porous layers under identical conditions, using similar strengths of AlCl3.6H2O and YCl3.6H2O at the same solution dipping time. Only different solution strengths of YbCl3.6H2O and ZrOCl2.8H2O were used for SM-1 and SM-2 preforms making runs. For SM-1 preform making run, we used higher solution strength of ZrOCl2.8H2O and lower strength of YbCl3.6H2O compared to SM-2 preform making run, maintaining the same solution strength of AlCl3.6H2O and YCl3.6H2O. After fabrication of preforms, thermal annealing was done at 1100°C for 3 hours in a controlled heating furnace at the heating and cooling rates of 20°C/min.
Fiber drawing from the annealed preforms was conventional. The fibers of 125 ± 0.5 µm diameter were drawn at a temperature around 2000°C using a fiber drawing tower. Both primary and secondary coating were applied on-line to increase the tensile strength as well as to reduce moisture ingress from outside. During drawing, a step proper control of fiber diameter, coating thickness, and coating concentricity along the whole its length were optimized for getting optical fibers of good quality. The thickness of primary coating (Desolite DP-1004) as well as secondary coating (Desolite DS-2015) and their uniformity were accomplished by adjusting the pressure of flow of the inlet gases into the vessel of primary and secondary coating resin during fiber drawing.
2. Phase separation, or immiscibility, is a phenomenon that exists in amorphous binary systems [18]. In bulk ZGA silicate glass, it is normally observed at temperatures below the onset of crystallization [19, 20]. However in some ZrO2–SiO2 systems, immiscibility exists even in the stable liquid phase above the melting point, while a stable immiscibility zone exists at high content of SiO2. Such stable zone extends to temperatures lower than the melting point and gives a meta-stable immiscibility zone in a wide composition range where phase separation occurs normally in an amorphous state [18], the circumstance accomplished in the present work. In such kind of glass, excess of aluminum does not prevent the formation of phase-separation as ZrO2 itself serves as a nucleating agent for phase-separation.
Notice that a proper choice of the doping levels of GeO2 and Al2O3 for fabricating Yb-doped silica fibers is important for reaching their optimal properties, including minimization of the background loss [21]. The fabricated Yb doped fibers (YFs) were optimized in this sense. The composition and the dopants average percentage together with their radial distributions in the preforms were measured by EPMA. The results for fiber preforms SM-1 and SM-2 are shown in Fig. 1 and in Table 1 . The average doping levels of GeO2 and Al2O3 are 2.0 and 7.0 mol% in both preforms, while those of ZrO2 / Yb2O3 are 1.25 / 0.75 mol% (SM-1) and 0.75 / 1.10 mol% (SM-2), correspondingly.
Fig. 1 Distributions of doping levels of GeO2 (a), Al2O3 (b), ZrO2 (c), and Yb2O3 (d) across the core area of optical fiber preforms SM-1 and SM-2 obtained from EPMA.
Table 1. Doping levels within core regions of preforms
The expected values of the doping levels of P2O5 and Y2O3 were around 0.025 and 0.05 mol%, respectively. Content of P2O5 was calculated from the gas phase reaction of POCl3 with O2 based on the vapor pressure of POCl3 at 25°C while content of Y2O3 was evaluated from the solution strength of YCl3.6H2O used within the solution. The role of phosphorous in our case was to enhance the phase-separation phenomena owing to the higher field strength difference (>0.31) [22] between Si4+ and P5+, not for suppressing PD (see below) because of its very low content.
3. We investigated further two fibers having different ratio of ZrO2/Yb2O3 to address the effects on nanostructuration means number of nano-particles along with their sizes within the core glass, which ultimately affect upon their spectroscopic and lasing properties. Here substitution of Y3+ with Yb3+ is negligible due to very low content of Y2O3. Moreover Warren and Pines [23] have pointed out that phase segregation is caused by the different structures formed due to the anions inside integrating with the cations with high field strength outside the structure of the amorphous network. The diameter of rare-earth ion (M3+) is larger than that of Zr4+. So the crystal field strength of Zr4+ ion is greater than that of rare-earth ions. The role of Y2O3 is to prevent cracking of the core glass. In the case of the Si–Al–Zr–O, Zr4+ and O2- ions are to combine to form Zr–O clusters where ZrO2 can be the nucleus. As a result of this, the number of the nano-particles formed along with their sizes increases with increasing the ratio of ZrO2/Yb2O3 as confirmed from TEM analyses.
A small section of thickness around 3 mm was cut down from the neck-down perform, which came out from the electrically controlled furnace of fiber drawing tower after heating at 2000°C before drawing of a fiber. Finally, the TEM specimens were mechanically polished and dimpled to thickness of about 10 μm. The final thinning of the samples to electron transparency was carried out using an Ar-ion mill. To evaluate the composition of phase-separated particles, the electron beam was focused on the particles and then focused in an area outside of the particles, when the energy dispersive X-ray analyses (EDX) data were taken. The nature of the particles was evaluated from their electron diffraction pattern. To make the TEM analyses of optical fiber samples, fiber coating was dissolved in acetone and stripped off to have bare fiber. After that the fiber was crashed in two different marble mortars and ground to fine powder; this glass powder was then dispersed in acetone liquid. Cu saver was rinsed into the liquid and took out. After drying, the glass powder was stick on the surface of the saver. The Cu saver was put under TEM to check the nano-particles.
TEM pictures for both the preform samples SM-1 and SM-2 are given in Figs. 2 and 3 , respectively, at some different positions of the whole core region. To discuss the mean size of the particles, we took the TEM pictures at the central region, core-clad boundary, and between the center and core-clad boundary for SM-1 and SM-2 samples.
Fig. 2 TEM pictures of the neck-down preform (SM-1) which comes out from the electrically controlled furnace of fiber drawing tower after its heating at 2000°C before drawing of fiber. Shown are the different areas (from right to left): (a) center of the core; (b) the region between core and core-cladding interface; (c) the core-cladding interface.
Fig. 3 TEM pictures of the neck-down preform (SM-2) which comes out from the electrically controlled furnace of fiber drawing tower after its heating at 2000°C before drawing of fiber. The different areas are shown from right to left: (a) center of the core; (b) the region between core and core-cladding interface; (c) the core-cladding interface. Inset – the electron diffraction pattern.
From the TEM pictures of the doping core of both samples, one can reveal the presence of phase-separated particles. Two to three TEM pictures taken at each position were used for calculating the average particle sizes. It is expected that separated phases ZrO2, Al2O3, and GeO2 in ZGA silicate glass trend to mix together during heating because under suitable doping levels a homogeneous amorphous mixture of the phases is thermodynamically more stable than for the separated ones. The nano-sized particles exampled by these figures are expected to be phase-separated areas rich in ZrO2, Al2O3, GeO2, and probably Yb2O3. The particles themselves were found to be amorphous (non-crystalline) in the nature as it is seen from the electron diffraction pattern shown on the segment shown in Fig. 3(c).
We didn’t check the homogeneity of the nano-particles distribution very precisely. The TEM pictures show that the numbers of nano-particles increases gradually from the center to the core-clad boundary. Though we didn’t measure the distribution of phosphorous content within the core region, the distribution profile of ZrO2 was estimated from EPMA, which shows almost uniform doping levels. It is expected that the doping levels of P2O5 will be increased from the center to the core-clad boundary region due to evaporation of P2O5 from the center core region during the collapsing stage at high temperatures, which is reflected in the refractive index profile of fiber SM-2 exampled by Fig. 4 .
Both ZrO2 [24] as well as P serves as nucleating agents for phase-separation. With increasing doping levels of ZrO2, the numbers of nano-particles will increase at each portion of the core region as well as from the center to the core-clad boundary region. The nano-particles formation in preform SM-1 at each portion of the whole core region having higher Zr/Yb ratio becomes higher than that of preform SM-2 having lower Zr/Yb.
The nano-particles formation in both preforms SM-1 and SM2 increases gradually from the center to core-clad boundary region. The sizes of nano-particles at three points of the core region of preform SM-1 becomes larger than those in preform SM-2 due to doping with larger content of ZrO2. As a result, the rate of nucleation for phase-separation of preform SM-1 is higher than that of preform SM-2. The average sizes of phase-separated particles is found to be 15 to 20 nm and 30 to 35 nm in preform samples SM-2 and SM-1, having Zr/Yb ratios of 0.68 and 1.66, respectively.
We have taken only one TEM picture of the fiber drawn from preform sample SM-1, see Fig. 5 . No major changes were observed from the viewpoint of sizes and number of nano-particles formed during drawing of the fiber from the preform. To make any conclusion about the effect of drawing process on the phase-separation, a more detailed study that takes several numbers of such TEM pictures under different drawing conditions would be required.
To evaluate the composition of phase-separated particles the electron beam was focused on the particles and in area outside the particles, when the energy dispersive X-ray analyses (EDX) data were taken. The nature of the particles was evaluated from their electron diffraction pattern. When the X-ray beam is focused on the particles, the signals (see Fig. 6 (b) ) associated with Zr, Al, Ge and Yb ions becomes higher than the signals coming out from outside the particles (see Fig. 6 (a)). These data signify that the particles are phase-separated ZGA-rich silicate. Similar results were observed for preform SM-1, in which the signal associated with Zr on the particles becomes higher than that shown in Fig. 6 (preform SM-2) due to doping of larger content of ZrO2.
Fig. 6 EDX spectra taken out (a) and on (b) of the particles in center of the core region (preform SM-2).
3. Final fibers characterization
We examine next the spectroscopic and laser properties of the final ZGA silicate YFs, also labeled SM-1 and SM-2, as being obtained from the correspondent preforms. The essential parameters of the fibers (NAs, inner and outer diameters, cutoff wavelengths, Yb3+ concentrations, and background loss) are given in Table 2 .
Table 2. Final fiber parameters
The fibers are easily and low-loss spliced with standard single-mode silica optical fiber components conventionally used in practice, owing to the similarity in the melting temperatures of the former and the latter [25].
Normally, the background loss should increase with increasing the sizes of the particles. The average sizes of phase-separated particles were found to be 15 to 20 nm and 30 to 35 nm of preform samples SM-2 and SM-1, having Zr/Yb ratios 0.675 and 1.66, respectively (see above). On the other hand, the doping level of Yb2O3 is much larger than ZrO2 in case of preform SM-2. So, the high background loss of fiber SM-2 seemingly arises due to the presence of large content of Yb2O3 though the sizes of the particles in this fiber are smaller than in fiber SM-1.
The absorption spectra of fiber samples SM-1 and SM-2 are presented in Fig. 7 . Experimentally, these spectra were obtained using a white-light source with a fiber output utilized as a probe beam and an optical spectrum analyzer (OSA) with a 0.5 nm resolution. The lengths of the fiber samples were chosen to be 140 (SM-1) and 100 (SM-2) cm.
Fig. 7 Attenuation spectra of SM-1 (red curve) and SM-2 (blue curve) YFs. Inset demonstrates the spectral details originated from other than Yb3+ features.
It is seen that the measured spectra are quite similar in the appearance for both fibers; a slight difference is observed only in extinction within the Yb3+ ions band around 976 nm (25 and 44 dB/m for SM-1 and SM-2, respectively) in regard to the background loss level within VIS to near-IR spectral range; the background loss correlates with the resonant Yb3+ peaks intensities. Interestingly, the zero phonon line at 976 nm in both fibers is measured to be about 5.5 nm only that is considerably less than in standard, zirconia free, alumino-silicate YFs; this is probably due to the lower interaction with the phonons system of zirconia based glass host [11]. The featured by the inset to Fig. 3 other spectral peculiarities (growing attenuation towards VIS) of the fibers can be provisionally regarded to Si or/and Ge related oxygen-deficient centers (below 600 nm) and non-bridged oxygen hole centers (NBOHC: 600–750 nm), well-known for zirconia free silica fibers [26–28].
We also found reasonable to compare the absorption (and fluorescence, see below) spectra of new ZGA based YFs and “standard” (nano-particles and zirconia free) germano-alumino silicate YFs, having comparable Yb3+ ions concentrations; see Fig. 8 . Figure 8(a) demonstrates the differences appeared in absorption (normalized on the peak value at 976 nm). Along with already discussed narrowing of the main (zero-phonon) peak, one can capture some spectral shifting of the other characteristic peak of Yb3+ (nearby 915–920 nm) in ZGA YF against the one of the standard Yb:Ge:Al YF. True reasons that could stand behind these observations need further studies.
Fig. 8 Attenuation (a) and fluorescence (b) spectra of SM-2 (blue curves 1) and “standard” Yb:Al:Ge (gray curves 2) YFs, having comparable contents of Yb3+ ions. All the spectra are normalized on the peak values of absorption and fluorescence (at 975-nm excitation).
The fluorescence kinetics of Yb3+ in the fibers was measured after in-core excitation at 975 nm wavelength using a standard laser diode (LD) with a fiber output and maximal output power of 300 mW. The LD output fiber was spliced to a tested YF that was angle-cleaved from the other side to prevent any feedback. The LD power was modulated by a driver controlled, in turn, by a function generator, to achieve square-shaped pulses of ms-width with sharp rise and fall edges. To avoid possible re-absorption, Yb3+ fluorescence was delivered from the lateral surface of the fiber sample by means of a multimode fiber. The time resolution of the experimental set-up was ~8 µs. The fluorescence decays shown in Fig. 9 were obtained after normalization of the fluorescence signals on their values registered at 10 µs after the pump switching off (zero time in the figures). Fiber samples’ lengths less than 5 cm were used in these experiments to avoid a contribution of unwanted amplified spontaneous emission (ASE), while the fluorescence signal was collected from a lateral surface of an YF sample at the point separated by approximately 5 mm from its splice with the LD output fiber. Typical Yb3+ fluorescence decays presented in Fig. 9 for SM-1 (a) and SM-2 (b) were obtained at pump power of 50 mW (i.e., well above the saturating powers for both YFs, which are about a few mW only; see below).
Fig. 9 Experimental fluorescence decay kinetics (symbols) of SM-1 (a) and SM-2 (b) YFs after excitation at 975-nm wavelength. Plain curves are single-exponent fits of the data.
It is seen that both fibers demonstrate virtually single-exponential fluorescence decay (measured, remind, after 10 µs after switching of pump light), with a time constant being τo = 0.79 ± 0.01 ms (the residual sum after fitting is better than R 2 = 0.98). The found values are comparable with those known for standard silica based YFs and seem to be promising for the laser/amplifying applications. Meanwhile, fluorescence decay was found to slightly depend on the content of Yb3+ dopants: The larger Yb3+ doping leads to a bit shorter fluorescence lifetime in the YFs. Also note that there was detected a shorter, scaled within a few microseconds, decay component in the fluorescence signal, which could not be resolved in our experimental arrangement; its nature remains uncertain and needs more precise measurements.
The fluorescence spectra of the fibers were obtained at excitation at 975 nm wavelength with the use of the same LD without modulation applied. In these experiments, we used a back-propagation detecting scheme (Fig. 10 ) instead of the lateral one, used in the fluorescence decay measurements, in order to increase fluorescence signals measured by OSA and, in the same time, to get true spectral features nearby the zero phonon line of Yb3+. A fiber sample was pumped through one of the two left ports of a 50/50 wavelength division multiplexer (WDM) while the other was connected to OSA. To prevent feedback caused by ASE and interference effects, an YF sample spliced to one of the two right ports of WDM and the other (free) right port of the latter were placed into special oil with high refractive index. We used in these experiments quite short lengths of YF samples, less than 10 (SM-1) and 5 (SM-2) cm.
The results of the experiments are demonstrated in Fig. 11 for fiber samples SM-1 (a) and SM-2 (b) and for different pump powers at 975 nm wavelength.
Fig. 11 Emission spectra (main frames) and saturation curves (insets) obtained for SM-1 (a) and SM-2 (b) YFs.
The fluorescence spectra obtained at the different pump levels are shown in main frames of the figures and the integrated powers within the Yb3+ fluorescence band are provided in insets to them. A few important consequences can be revealed from Fig. 11: (i) The fibers fluorescence spectra are generally similar to those of standard silicate YFs; (ii) low saturation powers are characteristic for both the fibers, Ps≈1.0…1.5 mW, which correspond to the saturating intensities Is≈30 kW/cm2.
Also refer to Fig. 8(b) for making direct comparison between the fluorescence spectrum of new ZGA YF and the one of standard Yb:Al:Ge YF. The definite features seen from the comparison could be a subject of more detailed studies.
Since at λ = 975–976 nm wavelength (the zero phonon line of Yb3+ ions) the absorption and emission cross-sections, σa and σe, are almost equal and the fluorescence lifetime τ 0≈0.8 ms (see Fig. 9), an estimate can be made for our ZGA YFs: σa≈σe = hc/2λτ 0 Is≈~5∙10−21 cm2 (where hc/λ≈2.0∙10−19 J is the energy quanta). The obtained cross-sections values are comparable with those known for other silica fibers doped with Ytterbium (see e.g. Ref. 29 and references therein). From here, we can estimate the average concentrations of Yb3+ dopants in the fibers (we disregard here the fact that they are really shared, see Fig. 2, between sub-ensembles of single, “dissolved” in the glass, Yb3+ ions and “clustered” ones within nano-particles): N0≈1.9 (SM-1) and 2.8 (SM-2) cm−3 (see also Table 2).
4. Laser experiments
The results of laser experiments with YFs SM-1 and SM-2 are highlighted by Figs. 12 –14 .
Fig. 12 Experimental setups for laser performance of YFs: Shown are the laser configurations with 100 & 4% couplers (a) and 100 & 10% couplers (b).
Fig. 14 Pump-power dependent lasing spectra of YF-based lasers obtained at the laser configuration with 100 & 4% couplers (a) and at the one with 100 & 10% couplers (b). Levels of pump power are given in insets to the figures.
The experimental arrangements are schematically shown in Fig. 12. A fiber sample was in-core pumped at 975 nm wavelength by the same fibered diode laser through a fibered isolator (not shown). Pump power, after passing the isolator and a narrow-band (0.2 nm) fiber Bragg grating (FBG) with high reflection (HR) coefficient R1 = 99.9% for 1058 nm was delivered to a piece of YF. The launched (into the YF) pump power was limited by around 280 mW. The laser cavity was formed by this and another FBG (with reflection coefficient R2 = 10% for 1058 nm) or by a perpendicularly cleaved YF end having non-resonant Fresnel reflection (~3.5%). In the experiments, different YF lengths were inspected but the results shown on the figures are obtained for the YF lengths of 17.8 (YF SM-1) and 11.6 (YF SM-2) m when most of the launched pump power was absorbed throughout the fiber while there were yet no pronounceable ASE noise.
These measurements were made using a standard power meter or photo detector. It was found that after crossing threshold (measured by 20–28 mW of pump power, for the different cavity implementations) the laser output linearly increases with the launched pump power. It is seen from Fig. 13 that for both YFs, SM-1 and SM-2, maximal laser release at 1058 nm wavelength is around 160 mW at 265 mW of pump power (at the use of Fresnel 3.5%-coupling), which corresponds to overall and slope efficiencies of about 60 and 70%, respectively. For shorter intra-cavity YF lengths, even higher efficiencies were obtained, up to 65 and 75%, correspondingly, but in that case more residual pump was observed on the laser output. As seen from Fig. 13, the lower reflectivity of the output FBG-coupler leads to the higher lasing efficiency, an expectable fact for the fibers with high gain, e.g. for YFs.
Fig. 13 Laser performances of YF-based lasers: Red (2,4) and blue (1,3) curves relate to SM-1 and SM-2 YFs, respectively. Curves 1 and 2 are for the laser configuration with 100 & 4% couplers and curves 3 and 4 – for the one with 100 & 10% couplers. Laser slope efficiencies are 64 (curve 1); 72 (curve 2); 29 (curve 3); 51 (curve 4) %.
Figure 14 examples the spectral features of lasing in the case of YF SM-1 (17.8 m).
The laser output spectrum was recorded using the same OSA, employed in the studies of YF absorption and emission spectra. It is seen that a narrow-line lasing (with bandwidth of around 0.35 nm defined mainly by the FBG spectral reflectivity) is released with a low-power ASE background. ASE noise looked spectrally a bit different in the cases of cavities formed by a pair of FBGs (R1,2 = 100 and 10%) and a couple of FBG (R1 = 100%) and Fresnel non-resonant reflection of a cleaved YF end. It is remarkable that the ratio of spectrally integrated output within the laser line (0.35 nm) and broad-band ASE (1025–1150 nm) in each case and for any pump power is of the order or less than 10−4, revealing excellent spectral characteristics of the lasers in terms of the noise (ASE) figure.
A very similar result has been obtained for SM-2 YF (not shown); a difference was a slightly higher level of ASE background for the laser configuration with a non-resonant Fresnel coupling.
5. Photo darkening (PD) measurements
Recently, the pump-induced PD in Yb3+-doped fibers has been recognized as a drawback for power-scaling of YF based lasers, especially in the case of heavily doped fibers [30]. It was found that the induced loss is proportional to the inversion level of Yb3+ ions. PD is most pronounceable in silicate YFs rather than in phosphate ones [30–35]. An inspection of how this effect manifests itself in the SGA silicate YF is of much interest. In the meantime, since the spectroscopic properties of a nano-particles based fiber imply a modification in the surrounding environment of Yb3+ ions, one could expect a different PD behavior in it.
First of all, note that the laser output power insignificantly suffered from degradation at long-term (hours) pump tests, revealing that the PD effect in the ZGA nano-particles based YFs is rather weak. However, we cannot confirm at the moment that this was a direct consequence of nano-engineered core glass, as it was the case in a recent work [10].
The results of PD measurements at no lasing conditions (i.e. at the use of short pieces of YF without reflectors), by applying the procedure described in detail in Ref [36], are highlighted below (Figs. 15 and 16 ). The effect of PD in the YFs was evaluated by monitoring the transmitted probe power at 633 nm wavelength (a He–Ne laser, 5 mW). The probe beam transmission was measured after sequencing doses of YF irradiation at 975 nm wavelength (at the moments when the pump light was temporally switched off). We used the same pump source as in the experiments discussed above. The LD output was spliced to one of the input ports of WDM while one of its output ports was spliced to the YF; the pump beam propagated through the YF and photo darkened it. The probe beam from the He–Ne laser was coupled to the YF through the second input port of WDM (so, it propagated in the same direction as the pump one). The attenuated output power at 633 nm was detected by a photo detector.
Fig. 15 Temporal behaviors of normalized transmitted power at 633 nm at CW pumping (975 nm, 255 mW) of SM-1 and SM-2 YFs.
Fig. 16 The examples of difference (PD loss) spectra obtained for SM-1 (a) and SM-2 (b) YFs after 10 and 90 min of 975-nm pumping.
We used in the experiments short pieces of YF samples: ≈30–35 (SM-1) and ≈20–25 (SM-2) cm, correspondingly, in order to have a possibility to measure the difference in the attenuation spectra (these spectra were obtained after some of the irradiation doses). Launched pump power at 975 nm was maintained in the experiments at 250 mW, providing a high Yb3+ population inversion that was virtually at the same level (>45%) throughout the YFs length.
The temporal characteristics of the transmitted power at 633 nm for SM-1 and SM-2 fibers are presented in Fig. 15 and the correspondent spectra differences as these appear through the PD tests are shown in Fig. 16.
It is seen that for the less Yb3+ doped fiber SM-1 the PD effect is almost negligible as compared to the heavier doped SM-2, an expectable fact. One can also reveal from the data that although the PD effect in new ZGA nano-particles based YFs is not pronounceable, its appearance resembles the PD features occurring in other alumino-silicate YFs. Only one atypical detail that stems from the PD loss dependences shown in Fig. 15 (see inset) should be emphasized: The transmitted power at 633 nm, at the initial PD stage, drops by a stepwise (not smooth [30–33,35,36]) manner, probably being the appearance of different particular rates of PD within and outside the nano-sized particles, correspondingly rich and poor in Yb2O3 content. However, this feature ought to be proved by more detail studies in the future.
6. Conclusions
New Ytterbium doped ZGA silicate nano-particles based optical fibers are fabricated through the conventional MCVD and SD processes. The start fiber preforms are studied by means of EPMA, EDX, and electron diffraction analyses, revealing phase-separated nano-sized Ytterbium-rich areas in their cores. The sizes and numbers of nano-particles formed in the host glass increase with increasing Zr/Yb ratio, which allows us to conclude that ZrO2 serves as a nucleating agent facilitating the formation of such kind phase-separated nano-particles. On the other hand, no major changes were observed in the parameters of the engineered nano-particles against the process of fibers drawing from the preforms. The final YFs are inspected in the sense of spectroscopic and laser properties along with the PD tests. The absorption and fluorescence (at 975-nm pumping) spectra of the fibers are characterized by a narrow (~5 nm) zero-phonon peak at 976 nm, which is narrower comparing the ones in standard Yb:Al:Ge co-doped silica fibers; there are also detected differences in the spectra as whole between the former and the latter fibers. A few important estimates are made for some important parameters of the ZGA silica YFs: Yb3+ fluorescence decays in both fibers with different Yb contents are measured by ~0.8 ms at 975-nm excitation while saturating powers / intensities are measured by ~1 mW / ~30 kW/cm2, respectively. Laser experiments with these fibers demonstrate high slope efficiency of lasing of ~75% and low noise (ASE) factor (~10−4), at in-core 975-nm pumping. The PD effect is revealed to be weak in the ZGA silica YFs, being in the meantime characterized by a peculiar stepwise behavior through the pumping time. The full characterization of the new ZGA preforms and YFs can serve a base for their use in the future for versatile laser and amplifier applications.
Acknowledgements
Authors acknowledge the DST (Government of India) and the CONACYT (Government of Mexico) for providing financial support through the Program of cooperation in science and technology between India and Mexico.
References and links
1. M. Fujii, M. Yoshida, S. Hayashi, and K. Yamamoto, “Photoluminescence from SiO2 films containing Si nanocrystals and Er: Effects of nanocrystalline size on the photoluminescence efficiency of Er3+,” J. Appl. Phys. 84(8), 4525–4531 (1998). [CrossRef]
2. J.St. John stJ. L. Coffer, Y. D. Chen, and R. F. Pinizzotto, “Synthesis and characterization of discrete luminescent erbium-doped silicon nanocrystals,” J. Am. Chem. Soc. 121, 1888–1892 (1999). [CrossRef]
3. C. E. Chryssou, A. J. Kenyon, T. S. Iwayama, C. W. Pitt, and D. E. Hole, “Evidence of energy coupling between Si nanocrystals and Er3+ in ion-implanted silica thin films,” Appl. Phys. Lett. 75(14), 2011–2013 (1999). [CrossRef]
4. W. T. Nichols, J. W. Keto, D. E. Henneke, J. R. Brock, G. Malyavanatham, M. F. Becker, and H. D. Glicksman, “Large-scale production of nanocrystals by laser ablation of microparticles in a flowing aerosol,” Appl. Phys. Lett. 78(8), 1128–1130 (2001). [CrossRef]
5. E. M. Yeatman, M. M. Ahmad, O. McCarthy, A. Martucci, and M. Guglielmi, “Sol-gel fabrication of rare earth doped photonic components,” J. Sol-Gel Sci. Tech. (Paris) 19, 231–236 (2000). [CrossRef]
6. G. Brasse, C. Restoin, J.-L. Auguste, S. Hautreux, J.-M. Blondy, A. Lecomte, F. Sandoz, and C. Pedrido, “Nanoscaled optical fibre obtained by the sol-gel process in the SiO2–ZrO2 system doped with rare earth ions,” Opt. Mater. 31(5), 765–768 (2009). [CrossRef]
7. M. Rajala, K. Janka, and P. Kykkänen, “An Industrial method for nanoparticle synthesis with a wide range of compositions,” Adv. Mater. Sci. 5, 493–497 (2003).
8. J. E. Townsend, S. B. Poole, and D. N. Payne, “Solution doping technique for fabrication of rare-earth doped optical fibres,” Electron. Lett. 23(7), 329–331 (1987). [CrossRef]
9. W. Blanc, B. Dussardier, and M. C. Paul, “Er doped oxide nanoparticles in silica based optical fibres,” Glass Technol.: Eur. J. Glass Sci. Technol. A 50, 79–81 (2009).
10. S. Yoo, M. P. Kalita, A. J. Boyland, A. S. Webb, R. J. Standish, J. K. Sahu, M. C. Paul, S. Das, S. K. Bhadra, and M. Pal, “Ytterbium-doped Y2O3 nanoparticle silica optical fibers for high power fiber lasers with suppressed photodarkening,” Opt. Commun. 283(18), 3423–3427 (2010). [CrossRef]
11. G. Brasse, C. Restoin, J.-L. Auguste, and J.-M. Blondy, “Cascade emissions of an erbium-yyterbium doped silica-zirconia nanostructured optical fiber under supercontinuum irradiation,” Appl. Phys. Lett. 94(24), 241903 (2009). [CrossRef]
12. D. E. Harrison, N. T. Melamed, and E. C. Subbarao, “A new family of self-activated phosphors,” J. Electrochem. Soc. 110(1), 23–28 (1963). [CrossRef]
13. C. Urlacher, J. Dumas, J. Serughetti, J. Mugnier, and M. Munoz, “Planar ZrO2 waveguides prepared by the sol–gel process: Structural and optical properties,” J. Sol-Gel Sci. Technol. 8(1-3), 999–1005 (1997). [CrossRef]
14. A. Patra, C. S. Friend, R. Kapoor, and P. N. Prasad, “Upconversion in Er3+:ZrO2 nanocrystals,” J. Phys. Chem. B 106(8), 1909–1912 (2002). [CrossRef]
15. V. C. Costa, M. J. Lochhead, and K. L. Bray, “Fluorescence line narrowing study of Eu3+ -doped sol-gel silica,” Chem. Mater. 8(3), 783–790 (1996). [CrossRef]
16. W. D. Kingery, H. K. Bowen, and D. R. Uhlmann, Introduction to Ceramics (Wiley, New York, 1976).
17. V. G. Zavodinsky and A. N. Chibisov, “Zirconia nanoparticles and nanostructured systems,” J. Phys: Conference Series 29, 173–176 (2006). [CrossRef]
18. P. F. James, “Liquid-phase separation in glass-forming systems,” J. Mater. Sci. 10(10), 1802–1825 (1975). [CrossRef]
19. G. D. Wilk, R. M. Wallace, and J. M. Anthony, “Hafnium and zirconium silicates for advanced gate dielectrics,” J. Appl. Phys. 87(1), 484–492 (2000). [CrossRef]
20. G. Rayner, R. Therrien, and G. Lucovsky, “The structure of plasma-deposited and annealed pseudo-binary ZrO2-SiO2 alloys,” Material Res. Soc. Symp. Proc, 611, C1.3.1–C1.3.9 (2000).
21. V. F. Khopin, A. A. Umnikov, N. N. Vechkanov, A. E. Rozental’, A. N. Gur’yanov, M. M. Bubnov, A. A. Rybaltovskii, A. V. Belov, and E. M. Dianov, “Effect of core glass composition on the optical properties of active fibers,” Inorg. Mater. 41(4), 434–437 (2005). [CrossRef]
22. M. Tomazawa and R. H. Doremus, eds., Phase Separation in Glass 17, Academic Press, NY, 1979, p. 71.
23. B. E. Warren and A. G. Pines, “Atomic consideration of immiscibility in glass systems,” J. Am. Ceram. Soc. 23(10), 301–220 (1940). [CrossRef]
24. P. Vomacka, O. Babushkin, and R. Warren, “Zirconia as a nucleating agent in a yttria-alumina-silica glass,” J. Eur. Ceram. Soc. 15(11), 1111–1117 (1995). [CrossRef]
25. M. C. Paul, S. W. Harun, N. A. D. Huri, A. Hamzah, S. Das, M. Pal, S. K. Bhadra, H. Ahmad, S. Yoo, M. P. Kalita, A. J. Boyland, and J. K. Sahu, “Performance comparison of Zr-based and Bi-based erbium-doped fiber amplifiers,” Opt. Lett. 35(17), 2882–2884 (2010). [CrossRef] [PubMed]
26. D. L. Griscom, “Trapped-electron centers in pure and doped glassy silica: A review and synthesis,” J. Non-Cryst. Solids 357(8-9), 1945–1962 (2011). [CrossRef]
27. E. M. Dianov, V. M. Mashinsky, V. B. Neustruev, O. D. Sazhin, V. V. Brazhkin, and V. A. Sidorov, “Optical absorption and luminescence of germanium oxygen-deficient centers in densified germanosilicate glass,” Opt. Lett. 22(14), 1089–1091 (1997). [CrossRef] [PubMed]
28. C. S. Carlson, K. E. Keister, P. D. Dragic, A. Croteau, and J. G. Eden, “Photoexcitation of Yb-doped aluminosilicate fibers at 250 nm: evidence for excitation transfer from oxygen deficiency centers to Yb3+,” J. Opt. Soc. Am. B 27(10), 2087–2094 (2010). [CrossRef]
29. P. Barua, E. H. Sekiya, K. Saito, and A. J. Ikushima, “Influences of Yb3+ ion concentration on the spectroscopic properties of silica glass,” J. Non-Cryst. Solids 354(42-44), 4760–4764 (2008). [CrossRef]
30. J. J. Koponen, M. J. Söderlund, H. J. Hoffman, and S. K. T. Tammela, “Measuring photodarkening from single-mode ytterbium doped silica fibers,” Opt. Express 14(24), 11539–11544 (2006). [CrossRef] [PubMed]
31. S. Yoo, C. Basu, A. J. Boyland, C. Sones, J. Nilsson, J. K. Sahu, and D. Payne, “Photodarkening in Yb-doped aluminosilicate fibers induced by 488 nm irradiation,” Opt. Lett. 32(12), 1626–1628 (2007). [CrossRef] [PubMed]
32. S. Jetschke, S. Unger, U. Ropke, and J. Kirchhof, “Photodarkening in Yb doped fibers: experimental evidence of equilibrium states depending on the pump power,” Opt. Express 15(22), 14838–14843 (2007). [CrossRef] [PubMed]
33. J. J. Koponen, M. J. Soderlund, H. J. Hoffman, D. A. Kliner, J. P. Koplow, and M. Hotoleanu, “Photodarkening rate in Yb-doped silica fibers,” Appl. Opt. 47(9), 1247–1256 (2008). [CrossRef] [PubMed]
34. Y.-W. Lee, M. J. F. Digonnet, S. Sinha, K. E. Urbanek, R. L. Byer, and S. Jiang, “High-power Yb3+-doped phosphate fiber amplifier,” IEEE J. Sel. Top. Quantum Electron. 15(1), 93–102 (2009). [CrossRef]
35. A.V. Shubin, M.V. Yashkov, M.A. Melkumov, S.A. Smirnov, I.A. Bufetov, and E.M. Dianov, “Photodarkening of aluminosilicate and phosphosilicate Yb-doped fibers,” CLEO-Europe – EQEC 2007, Technical digest: Paper CJ3–1-YHU (2007).
36. A. D. Guzman-Chávez, A. V. Kir’yanov, Yu. O. Barmenkov, and N. N. Il’ichev, “Reversible photo-darkening and resonant photo bleaching of Ytterbium-doped silica fiber at in-core 977-nm and 543-nm irradiation,” Laser Phys. Lett. 4(10), 734–739 (2007). [CrossRef]
