Atomic layer deposition (ALD) was used to fabricate an ytterbium (Yb)-doped silica fiber in combination with the
conventional modified chemical vapor deposition (MCVD) method. An MCVD soot-preform with a porous layer of SiO<sub>2</sub>
doped with GeO<sub>2</sub> was coated with layers of Yb<sub>2</sub>O<sub>3</sub> and Al<sub>2</sub>O<sub>3</sub> prior to sintering, using the ALD method. ALD is a surface
controlled CVD-type process enabling thin film deposition over large substrates with good thickness control, uniformity
and high conformality. A materials analysis study showed that the dopants successfully penetrated the full thickness of
320 μm of the soot layer. Preliminary preform and fiber experiments on refractive index profiles, background losses,
lifetime and the characteristic gain-loss curve were performed demonstrating the potential of this method for fabricating
Yb-doped fibers with high concentration of dopants.
By confining the Yb doping within a smaller radius in the center of the core of the few-mode large-mode-area fiber, the
fundamental mode, which overlaps better with the Yb ions, sees higher gain than the higher order modes, and dominates
the output. Hence, improved beam quality can be achieved. A confined-doped fiber with 41/395μm core/cladding
diameters is fabricated by Direct Nanoparticle Deposition (DND) process. The fiber is characterized in an amplified
spontaneous emission (ASE) source setup. Near-diffraction-limited beam quality (M<sup>2</sup>~1.3) is experimentally
We examine the temperature dependence of photodarkening in ytterbium-doped silica fibers. A sequence of consecutive photodarkening experiments are performed over the same fiber sample, which shows good repeatability with no apparent changes in the glass structure. We find that during infrared irradiation, the level of saturation of the losses can be determined by the fiber core temperature, independent of the previous state of photodarkening losses and fiber temperature, and also at low temperatures where the thermal bleaching is not activated. We observe that variations in the fiber core temperature, induced by pump absorption due to photodarkening, affect the inversion level and photodarkening processes. These effects in turn cause a discrepancy in determining the ion dependence. We highlight the importance of performing the experiments under isothermal conditions and we propose a new approach to control the fiber temperature at room temperature and at elevated temperatures. The approach is based on an isothermal Galinstan bath. The appropriateness of this method is shown by comparing it to different cooling methods, and the results are supported by simulations.
By using a modal interference method, photodarkening-induced refractive index change in ytterbium doped fibers is
measured with high accuracy. This refractive index change is positive at the ytterbium lasing wavelengths near 1080nm,
and it gradually approaches a saturated level over time. It is found that the refractive index change is linearly
proportional to the photodarkening-induced excess loss at an arbitrary probe wavelength in the visible band. The
maximum refractive index change in our experiment is close to 1×10<sup>-5</sup>. The origin and influence of the photodarkening-induced
refractive index change on fiber lasers is discussed.
We study thermal bleaching of photodarkening-induced loss in ytterbium-doped fibers. Post-irradiation heating
of a photodarkened fiber is shown to result in further increase the loss which is attributed to both a permanent
increase of loss-inducing color centers and a temperature-dependent broadening of the absorption spectrum. The
permanent heat-induced increase of loss is believed to indicate presence of an intermediate energy state in the
NIRpho tochemical mechanism for photodarkening. Further, we apply the demarcation energy curve approach
to derive the thermal activation energy of the induced defects. For the studied commercial 20-μm-core-diameter
LMA fiber, the energy distribution consists of a single peak, located at 1.3 eV with a FWHM of 0.31 eV.
We study the temperature of an Yb-doped large-mode-area (LMA) fiber during an accelerated photodarkening
experiment. In these measurements, photodarkening is optically induced by IR irradiation (i.e. 915 nm) while the fiber
temperature is measured by a thermal camera. Fiber temperature is observed to exceed 120 °C under conditions of 10.5
W of pump power and unforced air cooling. We show evidences that this temperature increase is caused by the lost
pump power due to photodarkening. A thermal model is used to explain the fiber temperature in terms of pump power
absorbed by photodarkening-induced defects. Furthermore, the effect of temperature on the rate of photodarkening and
saturation of the losses is studied. Both the photodarkening saturation level and the photodarkening rate are observed to
show significant temperature dependence that result on a variation of the photodarkening rate ion dependency. The use
of an air cooling system and low inversion measurements is shown to reduce the ion dependency from 7 to 4.5.
Photodarkening is a detrimental phenomenon known to affect ytterbium doped fibers. Methods to study the spectral and
temporal properties of the photodarkening induced loss were developed. The spectral shape of the photodarkening loss
measured from multiple aluminosilicate samples indicate that visible wavelength(s) could be used in benchmarking
fibers for their PD induced loss. Two principal methods, core and cladding pumping, were introduced to induce a known
and repeatable inversion to fiber samples. The photodarkening rate could be parameterized using a single variable,
inversion. More generally, the photodarkening rate was found to follow a simple power law and to be proportional to
[Yb]<sup>7±1</sup> (the excited state Yb ion density). Two methods, stretched exponential and bi-exponential, were used to fit the
rate measurements. Both fitting methods were found suitable, with the bi-exponential method having more potential in
increasing the understanding of the mechanism(s) behind photodarkening. Coiling induced spatial changes in the
inversion and subsequent photodarkening performance were demonstrated for a large-mode-area fiber laser.
A combined photodarkening and thermal bleaching measurement of a large-mode-area (LMA) ytterbium-doped fiber
(YDF) is presented. Photodarkened YDF sample is recovered to pre-photodarkened state by thermal annealing. As a
result, this approach enables repeated measurements with the same sample and therefore eliminates uncertainties related
to changing of the sample (such as sample length and splice losses). Additionally, our approach potentially improves the
accuracy and repeatability of the photodarkening rate measurement, and also allows automation of the measurement
Accuracy at which population inversion in an ytterbium-doped fiber can be determined by modeling is studied. Here inversion refers to the percentage of ions excited to a higher energy level by the various optical fields. Knowledge of rare-earth ion inversion is crucial for determining the photodarkening behavior of a fiber, but can also be used to study the gain and noise properties of amplifiers. Sample fibers are first characterized for their optical and mechanical properties (e.g. absorption and fluorescence spectrum, excited state lifetime, rare-earth concentration and geometrical dimensions). Fiber specific absorption and emission cross-sections are then derived from the measured fiber parameters. Two methods (i.e. McCumber theory and Fuchtbauer-Ladenburg relation) are used to determine the detailed shape and the relative level of absorption to emission at different wavelengths. A full numerical model is used to simulate both core and cladding pumped YDFs. In order to validate the inversion results produced by the simulator a comparison between the measured and simulated ASE spectra is made. Sensitivity of the simulated ASE spectrum on the different parameters is investigated. Uncertainty analysis is made to show the contributions of various measured parameters on the uncertainty of the inversion. The principal contributor of uncertainty on the inversion was found to be the cross-section values.
Development of photodarkening in two similar large-mode-area ytterbium doped fibers from different sources is compared. The excess loss induced by photodarkening is derived from transmission loss measurements of pristine and pumped or photodarkened samples. To accelerate the photodarkening process, cladding pumping is used so as to achieve high and uniform inversion through the sample. Further, intensity profiles are measured and compared in effort to detect
possible radial variations in the induced losses.