Open Access
Add to BibSonomy
Abstract
Ytterbium-doped strontium fluorosilicate optical fibers exhibiting intrinsically low optical nonlinearities were fabricated and characterized. Specifically, reductions up to ~1.5 dB, ~9 dB, and ~3 dB in Raman gain, Brillouin gain, and thermo-optic coefficients, respectively, were measured relative to conventional silica optical fibers. Additionally, fluorescence lifetime, and emission and absorption spectra for these fibers are presented and suggest enhanced performance relative to their more commonly employed aluminosilicate and phosphosilicate counterparts. Low quantum defect (<1.5%) operation in these fibers, coupled with their low thermo-optic coefficients, may ultimately yield high power fiber lasers with greater immunity to thermal-based parasitic processes. The results indicate the potential of these fibers and glass materials for high energy fiber-based applications.
© 2018 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
Optical nonlinearities limit power-scaling in high energy laser systems [1,2]. Amongst these nonlinearities, stimulated Brillouin and Raman scattering (SBS, SRS, respectively) [3], and nonlinear refractive index (n2) wave-mixing phenomena (e.g., Four-Wave-Mixing, Self-Phase Modulation) [4] strongly deteriorate laser properties once their respective critical power thresholds are exceeded [2,5]. Over the past two decades, the laser community principally has focused on the development of micro-structured large mode area (LMA) silica fibers, where the optical intensity per unit area is reduced by spreading the optical power over a larger effective area core. In addition to increasing the resultant complexity and cost of these fibers, such LMA designs introduce additional parasitic phenomena, such as transverse mode instability (TMI), which presently serves as the dominant limitation in further power-scaling [6,7].
In this work, a different approach is adopted to manage nonlinearities, namely attacking the parasitic effects through the enabling materials from which they originate (i.e., the glass fiber core) [8–10]. Indeed, the Brillouin gain coefficient (BGC), the Raman gain coefficient (RGC), the thermo-optic coefficient (TOC), and the nonlinear refractive index (n2) are influenced by material properties that can strongly affect the respective nonlinearity thresholds. As a testament of the success of this approach, reductions of ~20 dB and ~3 dB in BGC and RGC, respectively, for sapphire-derived and YAG-derived all-glass aluminosilicate fibers were demonstrated relative to conventional silica fibers [11,12]. However, if each nonlinearity can be individually reduced based on a given glass composition, it becomes quite motivating to consider the possibility of diminishing all nonlinearities concomitantly using a simple, conventional circular core-clad fiber geometry. Indeed, this is the purpose of this work. Specifically, this paper advances earlier efforts, where passive SrF2-Al2O3-derived aluminosilicate fibers, fabricated using the molten core method, were found to exhibit intrinsically low optical nonlinearities [10].
Here, active multicomponent fluorosilicate optical fibers were fabricated using the molten core method. Each fiber considered herein results from a different precursor composition, thus offering a broader range of core glass compositions for comparison of their respective properties. The BGC, RGC, and TOC values for these fibers are reported and discussed. Additionally, the active fibers were fabricated by incorporation of ytterbium (Yb) using either YbF3 or Yb2O3 as the precursor phase. Fluorescence lifetimes, emission and absorption spectra are also provided and discussed, along with preliminary lasing experiments.
2. Experimental procedure
2.1 Fiber fabrication
The fibers were fabricated using the molten core method [13]. Briefly, a material precursor, in the form here of a powder mixture, was inserted inside a pure silica glass capillary tube (3 mm inner and 30 mm outer diameters). This preform was drawn into fiber at a draw temperature (Tdraw), which was on the order of 2000°C. Upon heating, the precursor core phase melts, and some silica from the surrounding cladding is incorporated into the molten liquid precursor core. The high cooling rate (~2000 °C/s) experienced by the molten core during the draw process leads to a kinetically hindered glassy state in the resulting fiber. The targeted cladding diameter for all fibers was 125 µm and a conventional acrylate coating was deposited onto the fiber during the draw. Each drawn fiber was found to exhibit a circular core-clad geometry, with a graded refractive index profile for the core and was composed of the precursor materials along with silica. Approximately 800 m of fiber length was drawn for each sample.
The initial precursor composition and draw temperature for each fiber are provided in Table 1. The fibers’ designation in Table 1 informs the Reader as to the nature of the precursor employed in each case. Yb and YbF indicate that Yb2O3 or YbF3 were used, respectively, while SrAlSi and SrAlSiF correspond to initial SrO-Al2O3 and SrF2-Al2O3 powder mixtures. As an example, starting materials for YbF-SrAlSi fiber are YbF3, SrO and Al2O3. On the other hand, Yb-SrAlSiF fiber possesses the precursors Yb2O3, SrF2, and Al2O3. Additionally, the 23.75 and 71.25 mole% in Table 1 correspond to a 3 to 1 ratio between either SrF2 or SrO and Al2O3. This ratio was kept constant for consistency with previous work on SrAlSiF fiber [10].
Table 1. Draw temperatures and initial precursor compositions of the fabricated fibers.
2.2 Fiber characterization
Fiber refractive index profiles (RIP) were performed by Interfiber Analysis, LLC (8 Manns Hill Crescent, Sharon, MA 02067, USA). The RIPs were measured at a wavelength of 950 nm transversely through the side of the fiber using a spatially resolved Fourier transform interferometer [14]. Fortunately, dispersion in the refractive index difference over the wavelength range from 950 to 1550 nm is small for a large variety of glass systems, including aluminosilicates [15–17]. Therefore, since the systems investigated here are also aluminosilicates, the assumption is made that the dispersion characteristics of the present fiber RIPs are not markedly different. Consequently, the refractive index profiles measured at 950 nm are assumed to be identical to those in the 15XX nm region. This is important since several measurements at this wavelength were made, namely BGC, TOC, and optical attenuation, the latter being measured using the standard cut-back technique at a wavelength of 1534 nm. Typical cutback lengths were ~2 – 5 m.
Compositional analysis of the drawn fibers was performed using Energy Dispersive X-Ray (EDX) microprobe analysis on a HITACHI-6600 Scanning Electron Microscope (SEM), with an accelerating voltage of 20 kV. SEM micrographs were employed to characterize the shape and size of the core as well as to verify the absence of visible phase separation.
Spontaneous Raman spectra of the fiber core materials were obtained using a commercial Raman microscope (alpha300, WItech) in a backscattering geometry, utilizing a 532 nm pump source with a focused beam diameter of about 1 µm and a photon collection time of 120 s per datum. The Raman spectra are corrected similarly to that in Ref [18] and normalized relative to that for SiO2 (data point taken in the cladding, whose peak intensity value was set to 1). The Raman gain coefficient (RGC) for each fiber was then measured relative to SiO2 and corresponds to the intensity of the peak dominating the Raman bandwidth.
A heterodyne approach was employed to measure the Brillouin gain spectrum (BGS) of the fibers. The apparatus is similar to that in Refs [19–21]. In short, a pre-amplified narrow linewidth signal (∼100 kHz linewidth at 1534 nm wavelength) was launched into the fiber through an optical circulator. The back-reflected signal propagates through the circulator and is collected at the input of the fiber. Both Rayleigh and Brillouin back-scattered light are mixed on a fast PiN detector. The resultant electrical signal is probed by an electrical spectrum analyzer (ESA) revealing the Brillouin gain spectrum (BGS) of the fiber. Comparing the strength of this backscattered signal with that from a fiber of known BGC enables the measurement of the Brillouin gain coefficient (BGC). It is worth mentioning that the launched power is mode matched to the fiber under test in order to excite its fundamental mode. If only one Lorentzian peak is observed in the BGS (not including the one from the apparatus), this suggests there exists a single interaction between one optical and one acoustic mode (i.e., the one of the fundamental mode). Possible higher order acoustic and optical mode interactions can be observed in a BGS, taking the form of new frequency shifted scattering peaks or an asymmetry (non-Lorentzian) in the shape of the spectrum. However, none of these effects were observed in the fibers of this study, except for the fiber labeled YbF-SrAlSiF B. Other Brillouin-related properties such as the longitudinal acoustic velocity, the Brillouin frequency and linewidth, were also determined from the BGS.
The thermo-optic coefficient (TOC) of the fibers was measured using a fiber ring cavity laser discussed in greater detail elsewhere [21,22], in which the test fiber forms part of the cavity. As the temperature of the cavity fiber increases, the free spectral range (FSR) of the constructed laser is modified. The fiber (modal) dependence of the refractive index as a function of temperature can then be determined from a measurement of the FSR as a function of fiber temperature.
In order to measure fluorescence emission spectra, the fiber was spliced to a fiber-coupled semiconductor laser and pumped at 976 nm. A multimode patch cable, used to collect the light, is brought very close to the side of the fiber using a positioning mount. The patch cable was connected to an optical spectrum analyzer (OSA) and the spectrum subsequently registered. For fluorescence lifetime measurements, the same apparatus was used but the pump power was pulsed. The spectrum analyzer was set to a monochromator mode where the output was filtered to pass Yb luminescence. The signal then impinges on a silicon photon-counting avalanche photodiode where the decay data was collected. In order to measure the absorption spectrum, a short segment of the fiber (typically ~1 mm) was spliced between two passive fibers, and a broadband light source (white light) was passed through it. The spectrum was measured before and after passing through the fiber, and was divided to yield the absorption. Because cladding modes can be coupled into that short fiber segment, ‘ripples’ typically are observed in the absorption spectrum. The emission cross section is calculated using the Fuchtbauer-Ladenburg (F-L) formula [23,24]
where λ is the wavelength, τ is the fluorescence lifetime, n is the refractive index at the emission wavelength (λ), c is the velocity of light in the vacuum, I(λ) is the emission spectral intensity for a given wavelength, and the integral term corresponds to the total intensity of the emission bandwidth. Since the dispersion in the Yb luminescence band (~970 nm – 1100 nm) is expected to be small, the refractive index in Eq. (1) is taken to be independent of the wavelength for each fiber, and taken to be the value at the core center at the average emission wavelength (see section 3.6).3. Results and discussions
3.1 General fiber properties
The elemental compositions, in atomic percentage (At%) and taken at the core center of the fiber segments, along with other typical fiber parameters, are summarized in Table 2. The SrAlSiF fiber, previously characterized in [10], is reported for the benefit of the Reader. As the fiber is drawn, the precursor SrF2 melts and reacts with the SiO2, which dissolves into the core melt from the cladding. As a result, fluorine is lost through the formation of SiF4 and SrF2 is progressively transformed into strontium silicates and strontium aluminosilicates [25]. Similarly to SrF2, YbF3 is expected to progressively oxidize during fiber processing. The fluorine concentration ([F]) in Table 2 therefore originates from both residual SrF2 and YbF3 in the drawn fiber. Because the bonding nature of fluorine remains uncertain at present, the compositions in Table 2 are given in At% rather than in concentration of the oxide or fluoride compounds.
Table 2. Elemental compositions (at core center, in atomic percent, At%), core and cladding diameters (ϕcore and ϕclad, in µm), typical fiber parameters: fundamental LP0 mode index (n0) and mode area (Aeff, in × 10−12 m2), refractive index difference between core center and cladding (Δn, in × 10−3), and attenuation coefficient (α, in dB/m, at 1534 nm).
The core (center) / cladding (i.e., fused silica) index difference, Δn, is reported in Table 2 for each fiber. Since the Δn value factors in the modality of the fiber, which influences the effective nonlinearities, its material dependence in these fluorosilicate systems will be discussed. Materially, oxides of Sr, Al and Yb all increase the refractive index when added into silica [20,26,27], while F decreases it [28,29]. The effect of fluorine on the reduction of n is well-illustrated with fiber segment YbF-SrAlSiF B. This fiber exhibits the highest fluorine concentration at the core center ([F] = 5.26 At%) and the lowest Δn value, this even as this fiber contains higher Sr, Al, and lower Si levels relative to other fiber segments (e.g., Yb-SrAlSi). As high-power laser applications can require few-to-single mode operation, lower Δn values than the ones reported in Table 2 would be preferred. As already suggested in Ref [10], the possibility of using a pedestal design where an index-raising inner glass cladding layer is implemented between the core and silica cladding would reduce the fiber numerical aperture (NA) [30]. Additionally, the possibility of using higher refractive index glass cladding materials (e.g., in the SiO2-Al2O3-La2O3 system [31]) than SiO2 can also promote reduced Δn values.
In order to better understand the role of fluorine on the reduction of the refractive index of the glass core, a simple additivity model, described in [32], was employed. Here, the refractive index at the fiber core centers was calculated assuming only the presence of the oxide compounds: Yb2O3, SrO, Al2O3, and SiO2, similarly to the analysis performed in Ref [10]. Then, the difference between the calculated refractive index and the measured one from Table 2 constitutes the contribution of the fluorine. This diminution, normalized relative to the value for the all-oxide fiber (Yb-SrAlSi), as a function of fluorine concentration, is displayed in Fig. 1. Also reported in Fig. 1 is the diminution of the refractive index (at 1550 nm) for F-doped silica glasses taken from [28].
Fig. 1 Effect of fluorine to the diminution of the refractive index for multicomponent fluorosilicate glass optical fibers (in red) and for fluorine-doped silica glass (in blue, data taken from Ref [28]).
Interestingly, fluorine is found to reduce the refractive index similarly for both silica and fluorosilicate glasses over the range of compositions investigated. This result is somewhat surprising since it is known that fluorine can bond very differently in multicomponent glasses, for instance forming Al-F, Si-F, and Sr-F bonds [33–37] or more complicated complexes as described in Ref [38]. Consequently, one could expect fluorine to influence the glass refractive index differently depending on its (still to be determined) bonding nature. Additionally, the nature of the fluoride precursor from which the fluorine in the fiber core originates, i.e., SrF2, or YbF3, and the final fluorine concentration, do not influence the magnitude of refractive index reduction as a function of [F]. Further, the ratio between cationic species (i.e., Sr, Al, Yb, and Si) are different between the fibers but seem not to impact the effect that fluorine has on the reduction of n. Thus, it is possible that the bonding nature of fluorine is similar in all the fibers investigated.
Finally, it is worth mentioning that using the molten core method, fluorine concentrations that are 2 – 3 times higher than that typically permissible in conventional vapor-deposition silica optical fibers (~2 At%) are achieved [39], which enables the fabrication and investigation of novel glass compositions and determination of their properties.
As also presented in Table 2, the attenuation coefficients (α) were measured and found generally to be in the dB/m range at a wavelength of 1534 nm. While lower attenuation values will be critical for future applications of these fibers, these loss levels are acceptable for this materials study of composition, structure, and properties. Efforts are underway to develop fibers with reduced attenuation losses, ideally to below 50 dB/km, which includes starting with higher purity precursor materials. For completeness, the molten core method has yielded fibers with attenuation values in the 200 dB/m range [11,40]. Further, low-silica multicomponent optical fibers with losses of 20 dB/km have been successfully developed [41], and therefore fabrication of low loss multicomponent glasses is possible.
3.2 Optical nonlinearities
In Table 3, BGC, RGC, and TOC values for the characterized fibers are reported and compared to conventional silica optical fibers. The fibers exhibit intrinsically low nonlinearity coefficients, with reductions of 6-9 dB for BGC, 0.5-1.5 dB for RGC, and 1.2-3.2 dB for TOC, relative to conventional silica fibers. These results demonstrate the power of a unified materials approach to mitigating parasitic optical effects. The nonlinear refractive index values, n2, were not characterized for these fibers, but are estimated to be of similar magnitude (i.e., silica-like) to the SrAlSiF fiber [10]. Material contribution to the reduced nonlinear coefficients from Table 3 are discussed later in this work.
Table 3. Brillouin gain coefficient (BGC), Raman gain coefficient (RGC), and Thermo-optic coefficient (TOC) of multicomponent fluorosilicate optical fibers.
3.3 Raman spectroscopy
The normalized spontaneous Raman spectra for the Yb-doped fibers are shown in Fig. 2(a). The Raman spectrum for SiO2, taken in the pure silica cladding of one of the Yb-doped fibers, is provided for comparison. Typical features have long been identified for fused silica [42–44], including the principal Raman peak at ∼440 cm−1, associated with Si-O-Si stretching modes, and the peaks at ∼490 cm−1 and ∼600 cm−1 correspond to the so-called “defect” lines. Finally, peaks at ∼800 cm−1 (bending modes of Si-O bonds), 1060 cm−1, and 1200 cm−1 (associated with asymmetric and symmetric stretching vibrations) are observed. The presence of additional compounds in the glass serve to decrease the intensity of these SiO2 features, similarly to findings in Ref [10]. In silicate glasses, the characteristic modes at ∼1200 cm−1, ∼1100 cm−1, ∼950 cm−1, ∼900 cm−1 and, ∼850 cm−1, are associated with the formation of 0,1,2,3, and 4 non-bridging oxygen (NBOs) species per unit silica tetrahedron, respectively [42]. Interestingly, all the fibers exhibit somewhat identical features in this region, but with a range of magnitudes.
Fig. 2 a) Corrected and normalized spontaneous Raman spectra for Yb-doped fibers (and SiO2 as a reference). b) Relative Raman gain coefficient (RGC) for the Yb-doped fibers studied herein (in orange) as a function of SiO2 concentration. Complementary data points for optical fibers developed using the molten core method are reported to serve as comparison [9,10,45].
In strontium fluorosilicate glasses, it has been shown that fluorine preferentially bonds with the Sr [46,47]. Consequently, the presence of Sr-F bonds should be the source of the Raman peak at ∼275 cm−1 [48]. However, the strong overlap with the main silica peak renders its observation a challenge. In addition, its bandwidth broadens with temperature [48], diminishing its magnitude, and further complicates its observation.
The relative RGC for the fibers, already reported in Table 3, are displayed in Fig. 2(b) as a function of SiO2 concentration, along with other molten core fabricated aluminosilicate and silicate fibers previously characterized in Refs [9], [10], and [45]. For each fiber, the SiO2 concentration was calculated based on compositions in Table 2. In Fig. 2(b), it is clearly observed that all of these fibers follow the same general trend. This is somewhat expected as the reduction of RGC is proportional to the strength of the principal SiO2 scattering peak situated around 440 cm−1. Consequently, in these fibers where a reduction of ~25% in RGC relative to silica is typical, further increases in dopant concentrations (i.e., decreases in silica content) would yield concomitantly lower Raman response. In the limit case where a peak other than the main Si-O-Si peak starts to dominate the Raman bandwidth (e.g., in fibers with very high concentration of non-silica components), the trend in Fig. 2(b) will most likely not hold true.
3.4 Brillouin spectroscopy
The fibers presented herein all exhibit reduced BGC values, given in Table 3, from ~77% (6.3 dB) to ~88% (9.2 dB) relative to a conventional SMF-28 silica fiber. Other typical Brillouin-related properties, namely the Brillouin frequency shift (νB), the Brillouin linewidth (ΔνB), and the longitudinal acoustic velocity (Va), are reported in Table 4. For completeness, a representative Brillouin gain spectrum (BGS) for one of the fibers, YbF-SrAlSiF A, is reported in Fig. 3.
Table 4. Brillouin frequency shift (νB, in GHz), Brillouin linewidth (ΔνB, at 11 GHz and given in MHz), and longitudinal acoustic velocity (Va, in m/s) of the characterized fibers.
Fig. 3 Measured and normalized Brillouin gain spectrum (BGS) of the YbF-SrAlSiF A fiber segment (solid red line), at room temperature and zero-strain. The measured spectrum is fit with a Lorentzian curve (dashed black line). The peak situated at ~11 GHz is the signature of the SMF-28 used in the measurement apparatus.
Materially, the BGC is a function of several physical properties, and takes the following form [49]:
Here, n is the linear refractive index, p12 the transverse photoelastic coefficient (or Pockels coefficient), c the velocity of light in vacuum, λ the free space wavelength at which the measurement is performed (1534 nm), ρ the glass density, and ΔνB the Brillouin linewidth. The fabricated fibers exhibit large ΔνB values, from 2.8 to 4.2 times larger than ΔνB(SiO2), which, from Eq. (2), strongly contributes to the reduction of the BGC observed in these fibers. As SrF2 and YbF3 fluoride precursor materials largely oxidize during fiber processing, SrO and Yb2O3 are constituents in the final core glass. Since SrO, Al2O3 and Yb2O3 are known to increase the Brillouin linewidth [20,27], they contribute to an increase in ΔνB, hence reduction in BGC. Fluorine, on the other hand, has been shown to only slightly increase ΔνB [10]. These oxide species also exhibit negative p12 values (−0.245, −0.027, and −0.123, respectively, for SrO, Al2O3 and Yb2O3 [20,27]) when added to silica (positive p12, + 0.226 [20]). Hence, when mixed with silica, these compounds act to reduce the magnitude of the p12 glass value rendering a markedly reduced Brillouin response. They also increase glass density and Brillouin bandwidth, which further lessens the BGC relative to silica. The longitudinal acoustic velocities of the fiber cores in Table 4 are found to be lower than for the SiO2 cladding. Consequently, the fibers are acoustically guiding. Al2O3 increases Va relative to SiO2, whereas SrO and Yb2O3 decrease it. Therefore, the aggregate Va(fiber) results from a competing effect between these dopants, relative to silica.The contribution of F in the SrF2-derived core glasses is not yet fully understood [10]. Nevertheless, if one speculates that what fluorine remains is bonded to Sr as SrF2, crystalline values of SrF2 can be taken as a reference point to understand qualitatively how SrF2 in the glass could influence the BGC. SrF2 exhibits a density of 4240 kg/m3 [50], a positive p12 value of + 0.269 (at 633 nm) [50], and an acoustic velocity of ~5400 m/s [51]. With respect to previous discussions regarding the influence of dopant properties on the Brillouin response, the density value would favor a reduced BGC value. On the other hand, the p12 and Va values would promote an increase in BGC. In Table 4, the characterized fiber with the highest fluorine concentration (Yb-SrAlSiF B) also presents the lowest ΔνB value, even with higher Sr, Al, and Yb concentrations, which favors larger ΔνB values [20,27]. Comparatively, YbF-SrAlSiF, which contains only a minor amount of fluorine, exhibits a much larger ΔνB than the previous fiber, and suggests that a SrF2 glass compound reduces ΔνB, and, consequently, increases the BGC. Therefore, SrF2 is not expected to be any benefit in further reduction of BGC relative to SiO2. However, the partial oxidation of the fluorite precursor during fiber drawing promotes low Brillouin materials as SrO quite effectively reduces BGC when added to silica.
For completeness, these oxide dopants are known to increase n when added into silica. As this property also factors into BGC expression (Eq. (2)), maintaining a relatively low refractive index (i.e., close from silica) is important. In this context, as F is found to reduce the refractive index for these fluorosilicate glass cores relative to their pure oxide analogs, an increase of the BGC can be partially managed with SrF2 via lower n values.
3.5 Thermo-optic coefficient (TOC)
As shown in Table 3, the fibers exhibit reduced TOC values compared to conventional silica fibers. Figure 4 provides the evolution of the TOC as a function of fluorine content for the four Yb-doped fluorosilicate fibers characterized. The TOC of Yb-SrAlSi fiber, which contains no F, is also added as a point of reference. For completeness, TOC data for F-doped silica glasses, taken from Ref [28], also are displayed in Fig. 4.
Fig. 4 Thermo-optic coefficient (TOC) as a function of fluorine concentration (in atomic percent, At%) for both fluorosilicate fibers (in red) and F-doped silica glasses (in blue, taken from [28]).
The data show a generally linear relationship between TOC and fluorine concentration for both silicate and fluorosilicate families of glasses. Interestingly, the TOC is a strong function of [F], and the other dopants seem to play only a secondary role on the magnitude of TOC in these modified silicate glasses. Additionally, this reduction of TOC for the multicomponent fibers, relative to silica, is found for both fluorine containing fibers (Yb-SrAlSiF A and B, YbF-SrAlSiF A, and SrAlSiF) and the non-F-containing fiber (Yb-SrAlSi). From previous work on the SrO-Al2O3-SiO2 system [20], it was shown that SrO, as a glass constituent, contributes a negative value to the aggregate TOC value of the silicate core, as TOC(SrO) = −12.4 × 10−6 K−1, while Al2O3 slightly increases it with a TOC of 10.5 × 10−6 K−1. Therefore, TOC values for the fluorosilicate fibers are “shifted” to lower values, as they all contain SrO as a glass compound, although in differing proportions. Continuing the comparisons between the two families of glasses, it should be noted that fluorine decreases the TOC at a much larger rate in these fluorosilicates compared to F-doped silica glasses (−0.642 × 10−6 versus −0.287 × 10−6 K−1.[F]−1, respectively). The slope of the TOC with fluorine concentration is ~2.25 times more negative for the fluorosilicate glasses than for F:SiO2, suggesting a different bonding environment for the fluorine in these glasses (e.g., Sr-F instead of Si-F), but also clearly show the potential of these glasses for low thermo-optic response and high TMI threshold fibers.
By extrapolation of the linear fit in Fig. 4, a fiber with a zero TOC value can be achieved with ~12.4 At% of fluorine. If such a level of fluorine incorporation seems rather challenging, it should be mentioned that for practical purposes a reduction of 3 dB in TOC is already significant, and was achieved for the Yb-SrAlSiF B fiber (~3.25 dB reduction). Moreover, as previously discussed, conventional vapor-deposition methods limit F incorporation [39] to below this value. Thus, the Authors advocate the potential and interest in utilizing the molten core method in achieving such higher fluorine content compositions directly in fiber form.
For completeness, the TOC of crystalline SrF2 is negative, ~-13 × 10−6 K−1 from room temperature to 100 °C [52], and fluorine is known to decrease the refractive index upon heating [28]. Using Prod’homme’s derivation of the TOC [53], the low polarizability of the fluorine ion relative to oxygen, coupled with the large linear coefficient of thermal expansion (CTE) of SrF2 (18 × 10−6 K−1 at room temperature) drive TOC toward negative values. Therefore, and as observed, SrF2-derived glass compounds should reduce the magnitude of TOC relative to silica.
3.6 Ytterbium spectroscopy
Ytterbium (Yb) spectroscopic properties in these multicomponent fluorosilicate fibers are summarized in Table 5, along with the Yb and F concentrations ([Yb], [F]) at the fiber core centers. Normalized emission and absorption (cross section) spectra between the ytterbium 2F5/2 and 2F7/2 energy levels for some of the fabricated fibers are shown in Figs. 5(a) and 5(b).
Table 5. Yb and F fiber concentrations (at the core center, in atomic percent, At%), fluorescence lifetimes (τ, in µs), average emission wavelengths (λav, in nm), averaged emission cross sections (σem,av, in × 10−20 cm2) and performance figure of merit M, (M = σem,av × τ, in × 10−20 cm2.ms).
Fig. 5 a) Normalized emission cross section spectra for Yb-doped fibers. b) Normalized absorption cross section spectra for two fibers (Yb-SrAlSiF B and YbF-SrAlSiF A) relative to a commercial aluminosilicate laser fiber.
For fibers with high fluorine concentrations, the secondary peak in the emission bandwidth is shifted to shorter wavelengths and its magnitude increases relative to the dominant zero-phonon line peak. This behavior is similar to that found in Refs [54,55]. and is attributed to the nephelauxetic effect, where the ionicity of the active (Yb) ion in these fluorinated glasses modifies the spectroscopic properties and results in a blue shift with increasing fluorine concentration. The Yb3+ doping, through introduction of either Yb2O3 or YbF3 into an otherwise equivalent precursor composition, e.g., SrO-Al2O3 (black vs. purple curve in Fig. 5(a)), gives rise to a small, but measurable shift to shorter wavelengths. A similar behavior is also observed when Yb2O3 is substituted by YbF3 in an identical SrF2-Al2O3 precursor composition (blue vs. green curves in Fig. 5(a)). This clearly suggests some nearest-neighbor fluorine ions in the local environment of the Yb3+ ion. Remarkably, the non-fluorine-containing fiber (Yb-SrAlSi) experiences a blue-shift compared to commercial aluminosilicate fibers (commercial fiber in Fig. 5 but also in Ref [56]). This change in the emission bandwidth is attributed to the strontium also modifying the local environment of the ytterbium ion. Each of these observations are consistent with previous findings in strontium fluorosilicate glasses [47], where fluorine is found to be principally distributed around Sr and the rare earth (lanthanum in the study).
As observed in Fig. 5(b), both normalized absorption spectra of the YbF-SrAlSiF A and Yb-SrAlSiF B fiber segments exhibit a region of wide and very flat bandwidth, which resembles borate or phosphate glasses [54]. Furthermore, the zero-phonon line-strength of the YbF-SrAlSiF A fiber segment is considerably reduced compared to that of the Yb-SrAlSiF B segment. Interestingly, the two compositions have somewhat similar F concentration, but the starting precursor for YbF-SrAlSiF A segment is YbF3, whereas it is Yb2O3 for the Yb-SrAlSiF B. Since the line-strength of fluoride glasses is smaller than that for silicate glasses [54], it seems likely that some of the F remains proximate to the Yb3+ ions in these fibers when YbF3 is used as the precursor phase. Moreover, the zero-phonon line absorption peak is slightly shifted to shorter wavelengths for the YbF-SrAlSiF A fiber, relative to Yb-SrAlSiF B fiber, which is presumed to be due to a more highly fluorinated environment about the Yb3+ ions. For completeness, and as an example, absolute absorption and emission cross section spectra for the YbF-SrAlSiF A fiber is reported in Fig. 6.
Fig. 6 Absolute absorption and emission cross section spectra for YbF-SrAlSiF A fiber. The cross sections were determined using Eq. (1). The emission and absorption cross sections were set to be equivalent at the zero-phonon wavelength.
In Fig. 7 the average emission wavelength (λav), calculated from the equation reported in Table 5, is displayed as a function of the fluorine concentration. A fit to the data (λav = A + B/(C × [F] + D) is performed, and only serves as guide-to-the-eye; no physical meaning is attributed to it. The reduction of λav is noticeable even when only a few At% fluorine is present in the core glass, and becomes less pronounced with further additions until reaching an apparent asymptote. As will be seen in the following section, the observed blue shift is potentially beneficial for reduced quantum defect (QD) lasing, which is critical in high power fiber lasers where heat generation leads to TMI as well as other thermomechanical issues [2,7]. Moreover, lower emission wavelengths could also be beneficial for laser cooling applications [57].
Fig. 7 Average emission wavelength (nm) as a function of fluorine concentration (in atomic percent, At%).
An increase in radiative lifetime (τ), and decrease in both average emission wavelength (λav) and cross-section (σem,av), as a function of fluorine concentration, are characteristic of fluoride glasses [54]. On the other hand, a decrease in τ (and therefore increase in σem,av by means of Eq. (1)) with increasing Yb concentration is characteristic of lifetime quenching [58]. This effect is particularly visible for Yb-SrAlSiF B, in which its high fluorine concentration should promote longer lifetime. However, this fiber also exhibits the highest Yb content, resulting in a strong decrease of its τ value. Regardless, its upper state lifetime is still longer than that of the commercial fiber, even with nearly four times the Yb content. For completeness, it can be seen from Eq. (1) that τ scales inversely with n2. However, the decrease of n (~-0.005/[F] from Fig. 1) with fluorine content is expected to participate very little in the increase of τ. Finally, τ values up to 1.7 larger relative to the conventional aluminosilicate fibers (YbF-SrAlSiF A and B, Table 5) are observed.
The σem,av × τ product is used as a figure of merit for fiber laser performance [23,59]. This is especially true for a materials study such as this since both lifetime and cross-section values are material-dependent. One should note that, except for the Yb-SrAlSiF B fiber, which exhibits strong lifetime quenching, all the other fibers provide a better figure of merit (~25% to 50% higher) than the commercial aluminosilicate fiber, suggesting potentially enhanced performance of these multicomponent fluorosilicate fibers.
3.7 Fiber lasing
In order to demonstrate the feasibility of these multicomponent fluorosilicate fibers as laser gain media, a simple linear laser cavity was constructed from two custom-made fiber Bragg gratings (O/E-Land, LaSalle, Québec, Canada) with reflectivity values of 99.8% and 81.6% at 1010 nm, and two of the active fibers, YbF-SrAlSiF A (18 cm) and the commercial fiber described in Table 5 (20 cm). Figure 8 shows the measured output laser power (at 1010 nm) as a function of pump power (at 976 nm), represented by the data points. Modeling results using the theory in [60] are also displayed.
Fig. 8 Laser data for both YbF-SrAlSiF A and commercial Yb-doped aluminosilicate fibers, using pump wavelength of 976 nm, and output wavelength of 1010 nm.
Since the FBGs were written into separate Ge-doped fibers, cavity losses due to two splices, estimated via cutback to be ~0.3 dB and ~0.2 dB for the fluorosilicate and commercial fibers, respectively, were taken into consideration. In the case of the fluorosilicate fiber, background loss at the operating wavelength was determined to be roughly 1.36 dB/m, which was included in the simulation. The commercial fiber is assumed to be essentially lossless over its 20 cm length. The measured data and modeling results are in excellent agreement, with the slope efficiencies determined to be 34% and 53% for the fluorosilicate and commercial fibers, respectively. The moderate slope efficiencies for the two fibers result mainly from the splice losses, which is compounded by the high output reflectivity resulting in more passes through the cavity. Finally, had the fluorosilicate fiber been lossless, the slope efficiency is calculated to become 42%, and should further improve with a decreased output FBG reflectivity.
Work is underway to characterize these fibers in a more rigorously designed laser configuration and results are forthcoming. As it turns out, a QD below 1.5% has recently been demonstrated utilizing the YbF-SrAlSiF A fiber presented herein [61], using λinput = 981 nm and λoutput = 994.5 nm, and with a slope efficiency of 64%. The slope efficiency principally suffers from the somewhat high background and splice losses of the fiber. However, modeling suggests slope efficiencies >98% from this low QD configuration, provided that negligible attenuation values can be achieved. Therefore, these systems can be of consequent advantage over typical 5% QD conventional phosphosilicate and aluminosilicate fibers [62–64]. The potential for low QD operation in these fibers coupled with low TOC experienced by the optical mode can lead to further, scalable increases in the TMI threshold. Using YbF-SrAlSiF A fiber again as an example, its TOC value of ~5.14 × 10−6 K−1, combined with its reduced 1.5% QD, yields a TMI reduction of ~8.2 dB relative to a ~10.4 × 10−6 K−1 and 5% QD conventional fiber, everything else being set equal.
4. Conclusion
In this work, Yb-doped multicomponent fluorosilicate fibers were fabricated and their properties studied. Reductions of 6-9 dB, 0.5-1.5 dB and 1.2-3.2 dB of BGC, RGC, and TOC relative to conventional fibers were reported. It was shown that, materially, the addition of SrO and Al2O3 glass compounds in silica participates in the reduction of the BGC, principally through an increase of ΔνB and a decrease of p12. Additionally, the multicomponent nature of the glasses studied herein and their reduced silica concentrations yielded low Raman response materials. Finally, the incorporation of fluorine into the glass core principally contributes to the reduction of the TOC, resulting in fibers with higher resistance against parasitic thermo-optical effects (e.g., TMI, thermal lensing). Investigation of the spectroscopic properties of the Yb-doped fibers revealed the influence of fluorine on the lifetime and emission cross section of the fibers. Interestingly, the shift to lower wavelengths of the average emission wavelength is characteristic of the more ionic character of the oxyfluoride fibers relative to typical aluminosilicate systems. It is believed that these beneficial fluoride-glass-like spectroscopic properties also will ultimately lead to gainful control of thermal energy in high power fiber laser systems. As such, embodied in this glass family is the potential for laser fibers with both low-nonlinearity and low thermal impact, but united in assemblies with the robustness of silica fibers.
Funding
US Air Force Office of Scientific Research: Multidisciplinary University Research Initiative (MURI) “Internal Cooling of Fiber and Disc Lasers by Radiation Balancing and other Optical or Phonon Processes” (FA9550-16-1-0383); US Department of Defense High Energy Laser Joint Technology Office (HEL JTO) through the US Office of Naval Research, “A Unified Materials Approach to Mitigating Optical Nonlinearities in Optical Fiber” (N00014-17-1-2546).
Acknowledgments
The Authors also wish to thank Dr. Andrew Yablon (Interfiber Analysis) for the RIP measurements as well as Jianan Tang (Clemson University) for his help with the Raman spectroscopy.
References and links
1. M. N. Zervas and C. A. Codemard, “High Power Fiber Lasers: A Review,” IEEE J. Sel. Top. Quantum Electron. 20(5), 1–23 (2014). [CrossRef]
2. J. W. Dawson, M. J. Messerly, R. J. Beach, M. Y. Shverdin, E. A. Stappaerts, A. K. Sridharan, P. H. Pax, J. E. Heebner, C. W. Siders, and C. P. J. Barty, “Analysis of the scalability of diffraction-limited fiber lasers and amplifiers to high average power,” Opt. Express 16(17), 13240–13266 (2008). [CrossRef] [PubMed]
3. R. G. Smith, “Optical Power Handling Capacity of Low Loss Optical Fibers as Determined by Stimulated Raman and Brillouin Scattering,” Appl. Opt. 11(11), 2489–2494 (1972). [CrossRef] [PubMed]
4. R. H. Stolen and C. Lin, “Self-phase-modulation in silica optical fibers,” Phys. Rev. A 17(4), 1448–1453 (1978). [CrossRef]
5. A. V. Smith, B. T. Do, G. R. Hadley, and R. L. Farrow, “Optical damage limits to pulse energy from fibers,” IEEE J. Sel. Top. Quantum Electron. 15(1), 153–158 (2009). [CrossRef]
6. A. V. Smith and J. J. Smith, “Mode instability in high power fiber amplifiers,” Opt. Express 19(11), 10180–10192 (2011). [CrossRef] [PubMed]
7. L. Dong, “Stimulated thermal Rayleigh scattering in optical fibers,” Opt. Express 21(3), 2642–2656 (2013). [CrossRef] [PubMed]
8. J. Ballato, M. Cavillon, and P. Dragic, “A Unified Materials Approach to Mitigating Optical Nonlinearities in Optical Fiber. I. Thermodynamics of Optical Scattering,” Int. J. Appl. Glas. Sci. DOI 10.1111/ijag.12327 (2018).
9. J. Ballato and P. Dragic, “Materials development for next generation optical fiber,” Materials (Basel) 7(6), 4411–4430 (2014). [CrossRef] [PubMed]
10. M. Cavillon, C. J. Kucera, T. W. Hawkins, A. F. J. J. Runge, A. C. Peacock, P. D. Dragic, and J. Ballato, “Oxyfluoride core silica-based optical fiber with intrinsically low nonlinearities for high energy laser applications,” J. Lightwave Technol. 36(2), 284–291 (2018). [CrossRef]
11. P. Dragic, T. Hawkins, P. Foy, S. Morris, and J. Ballato, “Sapphire-derived all-glass optical fibres,” Nat. Photon. 6(9), 627–635 (2012). [CrossRef]
12. P. Dragic and J. Ballato, “Characterization of the Raman gain spectra in Yb:YAG-derived optical fibers,” Electron. Lett. 49(14), 895–897 (2013). [CrossRef]
13. S. L. Morris and J. Ballato, “Molten-core fabrication of novel optical fibers,” Am. Ceram. Soc. Bull. 92, 24–29 (2013).
14. A. D. Yablon, “Multi-wavelength optical fiber refractive index profiling by spatially resolved fourier transform spectroscopy,” J. Lightwave Technol. 28(4), 360–364 (2010). [CrossRef]
15. G. Ghosh, M. Endo, and T. Iwasaki, “Temperature-Dependent sellmeier Coefficients and Chromatic Dispersions for Some Optical fiber glasses,” J. Lightwave Technol. 12(8), 1338–1342 (1994). [CrossRef]
16. P. Dragic, M. Cavillon, and J. Ballato, “The linear and nonlinear refractive index of amorphous Al2O3 deduced from aluminosilicate optical fibers,” Int. J. Appl. Glas. Sci. DOI 10.1111/ijag.12337 (2017).
17. P. D. Dragic, C. Ryan, C. J. Kucera, M. Cavillon, M. Tuggle, M. Jones, T. W. Hawkins, A. D. Yablon, R. Stolen, and J. Ballato, “Single- and few-moded lithium aluminosilicate optical fiber for athermal Brillouin strain sensing,” Opt. Lett. 40(21), 5030–5033 (2015). [CrossRef] [PubMed]
18. M. D. O’Donnell, K. Richardson, R. Stolen, C. Rivero, T. Cardinal, M. Couzi, D. Furniss, and A. B. Seddon, “Raman gain of selected tellurite glasses for IR fibre lasers calculated from spontaneous scattering spectra,” Opt. Mater. 30(6), 946–951 (2008). [CrossRef]
19. P. Dragic, C. Kucera, J. Furtick, J. Guerrier, T. Hawkins, and J. Ballato, “Brillouin spectroscopy of a novel baria-doped silica glass optical fiber,” Opt. Express 21(9), 10924–10941 (2013). [CrossRef] [PubMed]
20. M. Cavillon, J. Furtick, C. J. Kucera, C. Ryan, M. Tuggle, M. Jones, T. W. Hawkins, P. Dragic, and J. Ballato, “Brillouin Properties of a Novel Strontium Aluminosilicate Glass Optical Fiber,” J. Lightwave Tech. 34, 1435–1441 (2016).
21. P. Law, A. Croteau, and P. D. Dragic, “Acoustic coefficients of P2O5-doped silica fiber : the strain-optic and strain-acoustic coefficients,” Opt. Mater. Express 1, 686–699 (2011). [CrossRef]
22. P. Dragic, M. Cavillon, and J. Ballato, “On the thermo-optic coefficient of P2O5 in SiO2,” Opt. Mater. Express 7(10), 3654–3661 (2017). [CrossRef]
23. J. Dong, M. Bass, Y. Mao, P. Deng, and F. Gan, “Dependence of the Yb3+ emission cross section and lifetime on temperature and concentration in yttrium aluminum garnet,” J. Opt. Soc. Am. B 20(9), 1975 (2003). [CrossRef]
24. M. J. Weber, D. C. Ziegler, and C. A. Angell, “Tailoring stimulated emission cross sections of Nd3+ laser glass: Observation of large cross sections for BiCl3 glasses,” J. Appl. Phys. 53(6), 4344–4350 (1982). [CrossRef]
25. M. Cavillon, “Molten core fabrication of intrinsically low nonlinearity glass optical fibers,” PhD Dissertation, Clemson University (2018).
26. J. E. Shelby, “Effect of morphology on the properties of alkaline earth silicate glasses,” J. Appl. Phys. 50(12), 8010–8015 (1979). [CrossRef]
27. P. D. Dragic, J. Ballato, S. Morris, and T. Hawkins, “The Brillouin gain coefficient of Yb-doped aluminosilicate glass optical fibers,” Opt. Mater. 35(9), 1627–1632 (2013). [CrossRef]
28. A. Koike and N. Sugimoto, “Temperature dependences of optical path length in fluorine-doped silica glass and bismuthate glass,” Proc. SPIE 6116, 61160Y (2006).
29. J. W. Fleming and D. L. Wood, “Refractive index dispersion and related properties in fluorine doped silica,” Appl. Opt. 22(19), 3102 (1983). [CrossRef] [PubMed]
30. P. Laperle, C. Paré, H. Zheng, and A. Croteau, “Yb-Doped LMA Triple-Clad Fiber for Power Amplifiers,” Proc. SPIE 6453, 645308 (2007). [CrossRef]
31. D. Litzkendorf, S. Grimm, K. Schuster, J. Kobelke, A. Schwuchow, A. Ludwig, J. Kirchhof, M. Leich, S. Jetschke, J. Dellith, J. L. Auguste, and G. Humbert, “Study of Lanthanum Aluminum Silicate Glasses for Passive and Active Optical Fibers,” Int. J. Appl. Glass Sci. 3(4), 321–331 (2012). [CrossRef]
32. P. D. Dragic, M. Cavillon, A. Ballato, and J. Ballato, “A unified materials approach to mitigating optical nonlinearities in optical fiber. II. A. Material additivity models and basic glass properties,” Int. J. Appl. Glas. Sci. DOI 10.1111/ijag.12328 (2018).
33. B. A. Shakhmatkin, N. M. Vedishcheva, and A. C. Wright, “Thermodynamic modelling of the structure of oxyhalide glasses,” J. Non-Cryst. Solids 345–346, 461–468 (2004). [CrossRef]
34. A. K. Swarnakar, A. Stamboulis, D. Holland, and O. Van Der Biest, “Improved Prediction of Young’s Modulus of Fluorine-Containing Glasses Using MAS-NMR Structural Data,” J. Am. Ceram. Soc. 96(4), 1271–1277 (2013). [CrossRef]
35. R. G. Hill, A. Stamboulis, and R. V. Law, “Characterisation of fluorine containing glasses by 19F, 27Al, 29Si and 31P MAS-NMR spectroscopy,” J. Dent. 34(8), 525–532 (2006). [CrossRef] [PubMed]
36. T. J. Kiczenski and J. F. Stebbins, “Fluorine sites in calcium and barium oxyfluorides: F-19 NMR on crystalline model compounds and glasses,” J. Non-Cryst. Solids 306(2), 160–168 (2002). [CrossRef]
37. C. Dalou, C. Le Losq, B. O. Mysen, and G. D. Cody, “Solubility and solution mechanisms of chlorine and fluorine in aluminosilicate melts at high pressure and high temperature,” Am. Mineral. 100(10), 2272–2283 (2015). [CrossRef]
38. B. O. Mysen and D. Virgo, “Interaction Between Fluorine and Silica in Quenched Melts on the Joins SiO2 - AlF3 and SiO2 - NaF Determined by Raman Spectroscopy,” Phys. Chem. Miner. 12(2), 77–85 (1985). [CrossRef]
39. K. Schuster, S. Unger, C. Aichele, F. Lindner, S. Grimm, D. Litzkendorf, J. Kobelke, J. Bierlich, K. Wondraczek, and H. Bartelt, “Material and technology trends in fiber optics,” Adv. Opt. Technol. 3, 447–468 (2014).
40. J. Ballato, T. Hawkins, P. Foy, B. Kokuoz, R. Stolen, C. McMillen, M. Daw, Z. Su, T. M. Tritt, M. Dubinskii, J. Zhang, T. Sanamyan, and M. J. Matthewson, “On the fabrication of all-glass optical fibers from crystals,” J. Appl. Phys. 105(5), 053110 (2009). [CrossRef]
41. S. Shibata, S. Mitachi, and S. Takahashi, “High numerical aperture multicomponent glass fiber,” Appl. Opt. 19(9), 1484–1488 (1980). [CrossRef] [PubMed]
42. P. F. McMillan, “Structural Studies of Silicate Glasses and Melts-Applications and Limitations of Raman Spectroscopy,” Am. Mineral. 69, 622–644 (1984).
43. F. L. Galeener, “Raman and ESR studies of the thermal history of amorphous SiO2,” J. Non-Cryst. Solids 71(1-3), 373–386 (1985). [CrossRef]
44. A. K. Yadav and P. Singh, “A review of the structures of oxide glasses by Raman spectroscopy,” RSC Advances 5(83), 67583–67609 (2015). [CrossRef]
45. P. D. Dragic and J. Ballato, “Characterisation of Raman gain spectra in Yb:YAG-derived optical fibres,” Electron. Lett. 49(14), 895–897 (2013). [CrossRef]
46. M. Środa and C. Paluszkiewicz, “The structural role of alkaline earth ions in oxyfluoride aluminosilicate glasses-Infrared spectroscopy study,” Vib. Spectrosc. 48(2), 246–250 (2008). [CrossRef]
47. M. Środa and Z. Olejniczak, “19F MAS-NMR studies of strontium oxyfluoride aluminosilicate glass,” J. Mol. Struct. 1001(1-3), 78–82 (2011). [CrossRef]
48. D. G. Mead and G. R. Wilkinson, “The temperature dependence of the Raman spectra of some alkaline earth crystals with the fluorite structure,” J. Phys. C Solid State Phys. 10(7), 1063–1072 (1977). [CrossRef]
49. G. P. Agrawal, Nonlinear Fiber Optics, 2nd ed. (Academic Press, 1995).
50. M. J. Weber, Handbook of Optical Materials (CRC Press, 2003).
51. K. C. Crouch, R. B. Rayment, and G. W. Marks, “Elastic Properties of Fluorides of Groups IA and IIA,” Final Rep. ZR 011 01 01 (NELC Z1), Nav. Undersea Warf. Center, DTIC Doc. # 848978 (1968).
52. A. Feldman, D. Horowitz, R. M. Waxler, and M. J. Dodge, Optical Materials Characterization Final Technical Report February 1, 1978 - September 30, 1978 (1978).
53. L. Prod’homme, “A new approach to the thermal change in the refractive index of glasses,” Phys. Chem. Glasses 1, 119–122 (1960).
54. M. Weber, J. Lynch, D. Blackburn, and D. Cronin, “Dependence of the stimulated emission cross section of Yb3+ on host glass composition,” Quantum Electron. IEEE J. 19(10), 1600–1608 (1983). [CrossRef]
55. Z. Lin, H. Zeng, Y. Yang, X. Liang, G. Chen, and L. Sun, “The effect of fluorine anions on the luminescent properties of Eu-Doped oxyfluoride aluminosilicate,” J. Am. Ceram. Soc. 93(10), 3095–3098 (2010). [CrossRef]
56. S. Suzuki, H. A. McKay, X. Peng, L. Fu, and L. Dong, “Highly ytterbium-doped silica fibers with low photo-darkening,” Opt. Express 17(12), 9924–9932 (2009). [CrossRef] [PubMed]
57. P. D. Dragic, M. Cavillon, A. Mironov, C. Kucera, T. Hawkins, J. G. Eden, and J. Ballato, “Yb-Doped Fluorosilicate Optical Fiber Development For Laser Cooling and Radiation Balancing Applications,” in Frontiers in Optics(p. JW4A.81), Optical Society of America (2017).
58. R. Paschotta, J. Nilsson, P. R. Barber, J. E. Caplen, A. C. Tropper, and D. C. Hanna, “lifetime quentching in Yb-doped fibers,” Opt. Commun. 136(5-6), 375–378 (1997). [CrossRef]
59. A. S. Pinheiro, A. M. Freitas, G. H. Silva, M. J. V. Bell, V. Anjos, A. P. Carmo, and N. O. Dantas, “Laser performace parameters of Yb3+ doped UV-transparent phosphate glasses,” Chem. Phys. Lett. 592, 164–169 (2014). [CrossRef]
60. G. Barnard, P. Myslinski, J. Chrostowski, and M. Kavehrad, “Analytical Model for Rare-Earth-Doped Fiber Amplifiers and Lasers,” IEEE J. Quantum Electron. 30(8), 1817–1830 (1994). [CrossRef]
61. N. Yu, M. Cavillon, C. Kucera, T. Hawkins, J. Ballato, and P. Dragic, “Low Quantum Defect Lasers via Yb-Doped Multicomponent Fluorosilicate Optical Fiber,” in Submitted to CLEO2018.
62. G. Gu, Z. Liu, F. Kong, H. Tam, R. K. Shori, and L. Dong, “Highly efficient ytterbium-doped phosphosilicate fiber lasers operating below 1020 nm,” Opt. Express 23(14), 17693–17700 (2015). [CrossRef] [PubMed]
63. S. Matsubara, K. Uno, Y. Nakajima, S. Kawato, T. Kobayashi, and A. Shirakawa, “Extremely low quantum defect oscillation of ytterbium fiber laser by laser diode pumping at room temperature,” in In Advanced Solid-State Photonics (P. TuB4). Optical Society of America (2007).
64. T. Yao, J. Ji, and J. Nilsson, “Ultra-low quantum-defect heating in ytterbium-doped aluminosilicate fibers,” J. Lightwave Technol. 32(3), 429–434 (2014). [CrossRef]
