We investigated photodarkening (PD) parameters of Yb/Al-doped silica fibers as a function of the concentration of additional rare earth ions like Tm or Er. It was found that both Tm and Er cause a decrease in Yb inversion followed by a reduction of PD in the case of Er, whereas Tm-codoping with more than 10 mol-ppm can strongly accelerate the process and also increase the PD loss. However, contrary to , we conclude that the typical PD behavior of Yb/Al fibers is an intrinsic feature of this fiber type and not caused by trace impurities of Tm (< 1 mol-ppm) unintentionally incorporated by the raw materials during fiber preparation.
©2011 Optical Society of America
Pump-induced photodarkening (PD) in Yb-doped silica fibers is a detrimental effect  well-characterized by accelerated measurements as a function of Yb inversion [3,4] and fiber temperature [5–7] but not fully understood because of the low energy of the photons and excited Yb ions involved, compared to the large energy gap up to the ultraviolet (UV) absorption levels that are assumed to be responsible for the effect [8, 9]. Atomic defects caused by the incorporation of Yb and Al in the SiO2 glass matrix are proposed to enable this process [10, 11]. The importance of both Yb excitation and pump or laser photons in the PD process was emphasized by modeling and laser experiments in .
A permanent optical attenuation was also observed in Tm-doped fibers exposed to 1064 nm radiation and explained by the stepwise absorption or cooperative upconversion processes . Codoping of Tm silica fibers with Yb was proposed to profit from energy transfer processes and gain improvement in fiber amplifiers by means of an auxiliary pump at 980 nm . However, enhanced PD in a Yb/Tm-codoped fiber was reported recently  and attributed to UV emission from Tm ions gathering up energy from excited Yb ions in a multi-step upconversion process. Because of the similar, though certainly three orders of magnitude weaker, Tm fluorescence detected in another Yb fiber contaminated with Tm by the raw material used in fiber preparation, it was concluded that even traces below 1 mol-ppm of Tm cause the whole PD process in Yb fiber lasers and explain the effect completely. However, this hypothesis was advanced from only two examined fibers.
In this paper, we investigate the influence of Tm ions on fluorescence properties and photodarkening parameters of Yb fibers; Tm concentration is varied in a wide range of more than four orders of magnitude. Two Yb/Er-codoped fibers are included in this study to compare the effect of Tm and Er on the PD process in Yb laser fibers.
Preform samples were prepared by MCVD (Modified Chemical Vapor Deposition) and solution doping according to a procedure with carefully controlled process steps described in . All preforms were collapsed in an O2/Cl2 atmosphere. For characterization purposes, fiber samples with a cladding diameter of 125 µm and a core diameter of approximately 10 µm were drawn. The fibers of the series listed in Table 1 were doped with nearly constant amounts of Yb2O3, Al2O3 and P2O5 but different contents of Tm2O3 varying from trace impurities (unintended contamination) to 550 mol-ppm. The fibers listed in Table 2 were chosen to study the effect of Er addition in fibers with comparable concentrations of the other dopants. The P content in these fibers is very low in contrast to commercial Yb/Er fibers that aim at an efficient energy transfer from pumped Yb to Er ions.
The preform samples were characterized by X-ray microprobe analysis (detection limit 20 mol-ppm; measurement error ± 6% for concentrations of > 100 mol-ppm). All concentrations specified in Tables 1 and 2 correspond to the maximum values of the measured profiles. Concentrations of Tm2O3 or Er2O3 below 10 mol-ppm were quantified from measurements of the core absorption in comparison to the other fibers of Tables 1 and 2 to an estimated accuracy of ± 50%. Trace impurities of Tm and Er were found to be generally < 0.2 mol-ppm in good consistency with the trace purity specified by the manufacturer of the raw material.
The fluorescence measurements were performed with an improved double-perpendicular technique (both the pump excitation at 980 nm and the detection of fluorescence light were carried out perpendicularly to the fiber axis ) to ensure identical measurement conditions and nearly saturated Yb inversion (≈0.4) for all fibers examined. Maximum measurement errors of ± 5 µs for the Yb lifetime and ± 5% for the fluorescence intensity were estimated. The measured fluorescence intensities were adjusted to take into account slight differences of core diameters and, in the case of Yb fluorescence intensity, also of Yb concentration.
The photodarkening kinetics was characterized at room temperature with core pumping and cladding pumping at 976 nm and 915 nm, respectively. A detailed description of the experimental set-up is given in . For cladding pumping, a tapered fiber bundle (TFB) coupler with signal feed-through was used instead of the WDM coupler. During pumping of short fiber pieces (1 to 2 cm), the core transmission T(t) relative to the value before PD was measured at a probe wavelength of 633 nm. The time dependence of the PD excess loss α(t) = −10⋅log(T(t))/L was calculated in units of dB/m (L is the fiber length), and the PD parameters were determined for the time interval of loss increase up to a potential maximum by a fitting procedure with the stretched exponential function described in .
The pump-induced Yb inversion was estimated from the pump power launched and the core material properties, by means of a commercial software (RP Fiber Power V2.0 ). The goal of our PD measurements was to compare all fibers for the same initiated Yb inversion of 0.46; therefore, a constant pump power of 200 mW for core pumping and 5.3 W for cladding pumping was applied. Only small differences of the density of excited Yb ions are caused by the actual Yb content of the compared fibers.
3. Results and discussion
In the core material of our examined fibers, an energy transfer from pump-excited Yb3+ ions to neighbored Tm3+ or Er3+ ions can take place . As a consequence, the envisaged Yb inversion is reduced, which manifests itself in a decrease of the measured Yb fluorescence intensity and lifetime. Both parameters are shown in Fig. 1 versus the concentration of Tm or Er. The expected effects become measurable if the Tm concentration exceeds 10 mol-ppm. Similar results were found for Tm- or Er-codoping at concentrations of several hundred mol-ppm. Therefore, comparable efficiencies of transfer to both types of rare earth ions can be assumed in our fibers analyzed (Tables 1 and 2).
Due to this energy transfer (and, in the case of Er, also to the direct excitation at 976 nm), the typical fluorescence peaks of the incorporated rare earth ions can be observed. Here, we mainly discuss the fluorescence caused by upconversion (UC) and excited state absorption (ESA), because we are interested in this path up to the UV range.
In Fig. 2(a) , typical Er3+ emission peaks in the visual range (VIS) are shown that are weak despite the relatively high Er concentrations in fibers #14 and #16 because of mostly nonradiative depopulation of the concerned Er3+ energy levels . Similar VIS fluorescence intensities are observed in a fiber with a Tm concentration of only 7 mol-ppm (Fig. 2(b)); obviously, UC processes are more efficient owing to the well-separated Tm3+ UC levels . A weak blue fluorescence (separated from the green Yb3+ cooperative luminescence around 500 nm) was also quantified in the fibers with Tm trace impurities. The intensity at 473 nm varying with the Tm concentration is shown in the inset of Fig. 2(b) for all fibers of the series. The signal first increases in proportion with Tm content, but weakens for higher values, probably because of the decreasing molar ratio of Yb/Tm. Further UC processes to UV levels of Tm3+, as described in , seem to be feasible.
The results of photodarkening measurements with core pumping at 976 nm are shown in Fig. 3 . For both Er-codoped fibers, the photodarkening loss is clearly reduced in comparison to fibers with only trace impurities (Fig. 3(a)). In contrast, Tm-codoping with more than 10 mol-ppm results in a strong acceleration of the photodarkening process and, for core pumping at 976 nm, also in a remarkable increase of the PD excess loss (Fig. 3(b)). In the fibers with highest Tm concentration, the PD loss decreases after passing a maximum. This interesting behavior was first observed at enhanced fiber temperatures  and explained by the existence of an intermediate state in the (thermal) bleaching path. Here, the bleaching could be forced by UV or VIS photons from Tm emission.
The results of the performed PD measurements are summarized in Fig. 4 by the presentation of PD parameters determined as functions of Tm concentration. The reproducibility of the PD parameters is better than ± 5% for the maximum loss and ± 20% for the rate constant. Larger fluctuations, especially for the fibers with only trace impurities, may be attributed mainly to slight differences of Yb and Al content caused by the complex process of fiber preparation. Nevertheless, it can be recognized that the PD effect is not enhanced in the range of more than two orders of magnitude of Tm concentrations up to 10 mol-ppm. In this range, the dotted lines approximate constant parameter values from a linear fit to the measurement values. General differences of PD parameters for pumping at 915 nm and 976 nm are observed despite equal Yb inversion and could be understood as a direct assistance of pump photons in the PD process as described in  because of the roughly four times higher pump intensity needed with core pumping at 976 nm to saturate the Yb inversion.
An obvious influence of Tm on the measured PD loss and rate constant is only observed for concentrations higher than 10 mol-ppm. A linear fit was applied to the experimental values in this range (solid lines in Fig. 4) to guide the eyes. Cladding pumping at 915 nm results in slightly decreased PD loss and moderately enhanced PD rates, whereas for core pumping at 976 nm both parameters are strongly increasing with the Tm concentration. Here, an impact of the pump wavelength is assumed and discussed below.
Actually, the reduction of Yb inversion (Fig. 2) by energy transfer to other rare earth ions is expected to diminish the PD effect, too. This behavior was found in the Er-codoped samples (Fig. 3(a)). In the Tm-codoped fibers, however, this effect is obviously compensated or even overcompensated (especially for core pumping at 976 nm) by another process as can be seen in Fig. 4 (the dotted lines describe the expected PD reduction estimated from the known dependencies of the PD parameters on the Yb inversion [3, 4]). This additional process seems to represent a Tm-assisted path to color center formation, probably enabled by UV photons emitted from Tm ions after excitation up to the UV level as it was suggested in .
Our measurements with core pumping at 976 nm confirm the remarkable increase of PD loss and rate in a fiber with thousand times the Tm content (about 100 mol-ppm in Fig. 4) compared to traces of Tm as discussed in . But the wide range of Tm concentrations examined by our fiber series allows us to differentiate between two different paths to PD, namely without and with the action of Tm. Certainly, the second path to photodarkening can take place in the vicinity of each Tm ion in Yb fibers. However, the number of emitted UV photons estimated in  for a fiber with trace impurities is extremely low (109 photons per second in a fiber length of 10 cm) and only partially initiates PD; many of these photons leave the fiber, whereas others may bleach existing color centers , resulting in an equilibrium state without a detectable Tm-assisted effect.
Moreover, the different pump wavelengths used in our experiments allow for a deeper understanding of the second path to PD regarding modeling and experimental results in . The first step of Tm excitation from ground state must be initiated by energy transfer from a neighbored Yb ion because no Tm fluorescence or photodarkening is provoked in Yb-free fibers by the pump wavelengths used; the next step can be enabled by another excited Yb ion or by excited state absorption (ESA) of a pump photon. The latter process is expected to dominate for pump intensities as used in our experiments and principally for increasing pump wavelengths because of better overlap with the ESA spectrum . This fact could explain the stronger increase in PD parameters in the case of pumping at 976 nm (Fig. 4).
In the Er-codoped fibers examined, only slight VIS fluorescence was measured, indicating that probably no efficient UC or ESA processes up to the UV level take place. So, the effect of Er ions is restricted to the reduction of both Yb inversion (due to a weak but existing energy transfer) and photodarkening even for doping with several hundred mol-ppm.
Our investigations of a series of Yb fibers with Tm addition varying in a wide range disprove the hypothesis of a superior role of Tm trace impurities in the PD process of Yb/Al fibers  but support the opinion that the prevailing effect is enabled by atomic precursors in the Yb/Al-doped silica core glass and driven by excited Yb ions and pump or laser photons.
The strong acceleration and enhancement of PD measured in Yb fibers with Tm-codoping of more than 10 mol-ppm is assumed to be caused by a second, Tm-assisted path of color center formation. This path of multi-step Tm excitation includes energy transfer from Yb ions and probably ESA of pump photons . The growing importance of ESA with increasing pump wavelength could be the reason why pumping at 976 nm affects the PD parameters of these fibers more strongly than pumping at 915 nm.
From our experimental results we conclude that photodarkening in Yb/Al silica fibers is an intrinsic feature of this core material. Trace impurities of Tm (below 1 mol-ppm) have no detectable impact on the parameters of this process.
References and links
1. R. Peretti, A.-M. Jurdyc, B. Jacquier, C. Gonnet, A. Pastouret, E. Burov, and O. Cavani, “How do traces of thulium explain photodarkening in Yb doped fibers?” Opt. Express 18(19), 20455–20460 (2010). [CrossRef] [PubMed]
2. 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]
3. S. Jetschke, S. Unger, U. Röpke, 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]
5. M. J. Söderlund, J. J. Montiel i Ponsoda, J. P. Koplow, and S. Honkanen, “Heat-induced darkening and spectral broadening in photodarkened ytterbium-doped fiber under thermal cycling,” Opt. Express 17(12), 9940–9946 (2009). [CrossRef] [PubMed]
6. S. Yoo, A. J. Boyland, R. J. Standish, and J. K. Sahu, “Measurement of photodarkening in Yb-doped aluminosilicate fibres at elevated temperature,” Electron. Lett. 46(3), 233–244 (2010). [CrossRef]
7. M. Leich, S. Jetschke, S. Unger, and J. Kirchhof, “Temperature influence on the photodarkening kinetics in Yb-doped silica fibers,” J. Opt. Soc. Am. B 28(1), 65–68 (2011). [CrossRef]
8. 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]
9. M. Engholm, L. Norin, and D. Aberg, “Strong UV absorption and visible luminescence in ytterbium-doped aluminosilicate glass under UV excitation,” Opt. Lett. 32(22), 3352–3354 (2007). [CrossRef] [PubMed]
11. K. E. Mattsson and J. Broeng, “Photo darkening of ytterbium cw fiber lasers,” Proc. SPIE 7195, 71950V, 71950V–9 (2009). [CrossRef]
12. M. M. Broer, D. M. Krol, and D. J. Digiovanni, “Highly nonlinear near-resonant photodarkening in a thulium-doped aluminosilicate glass fiber,” Opt. Lett. 18(10), 799–801 (1993). [CrossRef] [PubMed]
13. J. Chang, Q.-P. Wang, and G.-D. Peng, “Optical amplification in Yb3+-codoped thulium doped silica fiber,” Opt. Mater. 28(8-9), 1088–1094 (2006). [CrossRef]
14. J. Kirchhof, S. Unger, A. Schwuchow, S. Grimm, and V. Reichel, “Materials for high-power fiber lasers,” J. Non-Cryst. Solids 352(23-25), 2399–2403 (2006). [CrossRef]
15. A. Schwuchow, S. Unger, S. Jetschke, and J. Kirchhof, “Advanced methods for analyzation of absorption and fluorescence characteristics of rare-earth-doped silica,” in preparation.
18. F. E. Auzel, “Materials and Devices Using Double-Pumped Phosphors with Energy Transfer,” Proc. IEEE 61(6), 758–786 (1973). [CrossRef]
19. I. Manek-Hönninger, J. Boullet, T. Cardinal, F. Guillen, S. Ermeneux, M. Podgorski, R. Bello Doua, and F. Salin, “Photodarkening and photobleaching of an ytterbium-doped silica double-clad LMA fiber,” Opt. Express 15(4), 1606–1611 (2007). [CrossRef] [PubMed]
20. D. A. Simpson, W. E. Gibbs, S. F. Collins, W. Blanc, B. Dussardier, G. Monnom, P. Peterka, and G. W. Baxter, “Visible and near infra-red up-conversion in Tm3+/Yb3+ co-doped silica fibers under 980 nm excitation,” Opt. Express 16(18), 13781–13799 (2008). [CrossRef] [PubMed]