Spectral Measurement and Characterization of the EDZF

61 separated phases ZrO2 and Al2O3 will tend to mix into a homogeneous mixture before crystallization.

In addition to the physical structure of the fibre, the concentration levels of the various dopants in the EDZFs are analyzed through Electron Probe Micro-Analysis (EPMA). The obtained results from the EPMA are given in Table 2.

Table 2: Doping levels within core region of the preforms

It can be seen from the results of the EPMA given in Table 2 that the A and ZEr-B contain almost similar amounts of Al2O3 with dopant concentrations of between 0.24-0.25 mole%. However, the ZrO2 dopant concentration in the ZEr-B fibre is higher, at 2.21 mole% as compared to only 0.65 mole% in the ZEr-A fibre. Similarly, the Er2O3 dopant concentrations for ZEr-A and ZEr-B are 0.155 and 0.225 mole%

respectively, as is to be expected.

The next section will detail the measurement and characterization of the spectral properties of the EDZF.

62 providing real-time measurement and computation of the spectral losses at any wavelength in dB/Km. Figure 19 provides the schematic setup of the system.

Figure 19: Setup of the Bentham spectral attenuation measurement system

The system uses a 10 W quartz halogen light source with an aspheric lens as the primary light source for the system. The light source is powered with a constant 8.5 A current, and provides a stable output with fluctuations of less than +0.1% over an 8-hour period. The monochromator system consists of Czerny Turner optical

InGaAs PHOTODIODE

MONO-CHROMATOR

LOCK-IN AMPLIFIER

PERSONAL COMPUTER STEPPER MOTOR DRIVE

(PMC3B/IEEE)

FIBRE CHOPPER

LIGHT SOURCE

POWER SUPPLY

LAUNCH OPTICS

I/O commands and data stream

Electronic signal / current Optical signal / light beam Notation on Signals:

CURRENT AMPLIFIER

PSD

ADC

63 configuration which has a 300 mm focal length mount with 69 mm x 69 mm plain diffraction gratings, variable slits, adjustable from 0.01 mm to 5.5 mm and a stepper motor for controlled wavelength selection. An optical chopper is placed between the quartz light source and the monochromator. The monochromator is controlled by a PMC3B/IEEE¹⁷ stepper motor driver, which in turn is controlled by a personal computer. The output of the monochromator is connected to the launch optics, which includes a precision x-y-z positioner¹⁸, which will focus the light into the fibre under test. The fibre being tested is held in a bare fibre chuck capable of holding fibres with diameters from a few microns up to a maximum of 250 m. An Indium-Galium-Arsenide (InGaAs) photodiode with a 2 mm housing diameter is connected to the fibre chuck together with an adaptor. The LS4A lock-in-amplifier used in this setup uses a programmable current preamplifier, connected to a voltage sensitive lock in the module as well as an integrated analog to digital converter. The integrating analog to digital converter digitizes the lock-in output and passes it to the computer, allowing for the maximum use of the signal information available and for the computer to conduct digital averaging straight away. The software controlling the lock-in amplifier is based in a personal computer, which also controls the entire system via the IEEE/488 General Purpose Interface Bus (GPIB) interface.

The operation of the spectral measurement system begins with the selection of the starting and finishing wavelengths, as well as the spectral resolution and signal averaging period. Automatic zero routines will address any drift in electronic offsets by making a zero level measurement prior to each spectral run. The modulated white light is then allowed to pass through the monochromator before being launched into one end of the fibre under test. The light emerging from the other end of the fibre is collected by the large area InGaAs diode and the resulting signal amplified by a lock-in-amplifier. The spectral curve is obtained from the lock-in amplifier is then computed and stored by the computer. The fibre at the detector end is then taken out and a short length is selected, cleaved and reinserted in the detector assembly, after which a new spectral curve measurement is run and the obtained data stored.

The spectroscopic properties such as absorption coefficient, fluorescence and fluorescence decay curves of the fabricated fibres can be measured by the system.

When placed under test, the ZEr-A fibre shows a peak at 980 nm, with a loss of about

17 IEEE designates that the driver can be controlled by an IEEE/488 General Purpose Interface Bus (GPIB) 18 Adjustments of the x-y-z positioner are carried out manually.

64 16.5 dB/m, as well as a second peak at 1550 nm with a spectral attenuation of around 27.5 dB/m. The ZEr-B fibre sample also shows two attenuation peaks, the first occurring at 980 nm, with a loss of about 22.0 dB/m, while the second occurs at 1550 nm, with a loss of around 53.0 dB/m. These results are as expected, as the spectral attenuation curve of the fibre will be heavily influenced by the concentration of Erbium ions within the fibre. It is worthwhile to note that both fibres also show a peak at 800 nm, approximately 6.0 dB/m for the ZEr-A and 8.0 dB/m for the ZEr-B. The spectral attenuation curves of both fibres are shown in Figure 20.

Figure 20: Spectral attenuation curve of ZEr-A (above, left) and ZEr-B (above, right)¹⁹

The fluorescence spectra of the fibre samples were measured by laterally pumping, the fibres at 980 nm and a power of 100 mW. The fluorescence curves for both fibres are shown in Figure 21, while Figure 22 shows the fluorescence decay curves of both.

19 Figure modified from that in H. Ahmad, M. C. Paul, N. A. Awang, S. W. Harun, M. Pal and K. Thambiratnam, "Four-Wave-Mixing in Zirconia-Yttria-Aluminum Erbium," J. Europ. Opt. Soc. Rap. Public., vol. 7, pp. 12011-1 - 12011-8, 2012.

65

Figure 21: Fluorescence curves of the (a) ZEr-A and (b) ZEr-B fibres at a pump power level of 100 mW²⁰.

Figure 22: The fluorescence decay curves of the (a) ZEr-A and (b) ZEr-B fibres at a pump power level of 100 mW²¹.

20 Figure modified from that in H. Ahmad, M. C. Paul, N. A. Awang, S. W. Harun, M. Pal and K. Thambiratnam, "Four-Wave-Mixing in Zirconia-Yttria-Aluminum Erbium," J. Europ. Opt. Soc. Rap. Public., vol. 7, pp. 12011-1 - 12011-8, 2012.

21 Figure modified from that in H. Ahmad, M. C. Paul, N. A. Awang, S. W. Harun, M. Pal and K. Thambiratnam, "Four-Wave-Mixing in Zirconia-Yttria-Aluminum Erbium," J. Europ. Opt. Soc. Rap. Public., vol. 7, pp. 12011-1 - 12011-8, 2012.

1450 1500 1550 1600 1650

-100 -90 -80 -70 -60

Emission spectral power density, dBm

Wavelength(nm) a

b

0 5 10 15 20 25 30 35

-2.2 -2.0 -1.8 -1.6

-1.4 ^{y0 -0.00213 ±1.5999E-6}

A1 0.00069 ±1.9087E-6 t1 10.9349 ±0.08895

Detector, V

Time, ms

0 5 10 15 20 25 30 35

-2.2 -2.0 -1.8 -1.6

-1.4 _{y0 -0.00220 ±7.3068E-7}

A1 0.00082 ±8.804E-7 t1 10.86282 ±0.03404

Time, ms

Detector, V

(a) (b)

66 It can be seen from the figures that the two fibres, ZEr-A and ZEr-B, show almost the same fluorescence life-times of 10.93 and 10.86 ms respectively. Fibre ZEr-B however, which has higher doping levels of Er2O3 and ZrO2, shows a slight shorter fluorescence life-time. These results indicate that the concentration-quenching phenomenon, which is typical associated with high Er³⁺ ions concentrations, is strongly reduced through an increase in the doping levels of ZrO2.