Combination of optical µµµµ -transmission and µµµµ -
photoluminescence techniques for local characterization of rare earth doped glass µµµµ -spheres
D. Navarro-Urrios,
1,2M. Baselga,
1F. Ferrarese Lupi,
1L. L. Martín,
3C.
Pérez-Rodríguez,
3I. R. Martín,
3C. Vasconcelos,
4and N. E. Capuj
51 MIND-IN2UB, Dept. Electrònica, Universitat de Barcelona, C/ Martí i Franquès 1, 08028 Barcelona, Spain
2 Catalan Institute of Nanotechnology (CIN2-CSIC), Campus UAB, Edifici CM3, 08193 Bellaterra, Spain
3 Departamento de Física Fundamental y Experimental, Electrónica y Sistemas and MALTA Consolider Team, Universidad de La Laguna, 38206 Tenerife, Spain
4 Department of Technological Sciences and Development, Campus de Ponta Delgada, Azores University, 9501-801 Ponta Delgada, Açores, Portugal
5 Departamento Física Básica, Universidad de La Laguna, 38206 Tenerife, Spain
Outline
• Introduction and motivation
• Experimental details
• µ -transmission and µ -PL results
• Spectral dependence of the Q-factors
• Conclusions
Outline
• Introduction and motivation
• Experimental details
• µ -transmission and µ -PL results
• Spectral dependence of the Q-factors
• Conclusions
Introduction
840 860 880 900 920
Intensity (arb. units)
Wavelength (nm)
Effect of modulation in the emission of the Nd
3+:
4F
3/2→
4I
9/2produced by the microsphere.
840 860 880 900 920
Intensity (arb. units)
Wavelength (nm)
4
0 100 200 300 400 500 600
860 870 880 890 900 910 920
Transmission
Transmitted Intensity (arb.units)
Wavelength (nm)
TM polarisation
Photoluminescence
PL Intensity (arb.units)
TR polarisation
0 100 200 300 400 500 600
R=9.5 µ m
Motivation
A. Ashkin and J.M. Dziedzic. Applied Optics 20 (1981) 1803-1814
Combine transmission and PL measurements on light emitting micro- spheres because of their complementarity.
It is unlikely to produce perfect spheres, most probably obtaining spheroid shapes → WGM spectral positions depend on the specific normal vector angle of the optical trajectory.
Solution:Reduce the region of observation and share the collection scheme for both techniques.
Motivation
Outline
• Introduction and motivations
• Experiment details
• µ -transmission and µ -PL results
• Spectral dependence of the Q-factors
• Conclusions
Sample’s production
• Precursor material: Nd
3+-doped bulk borate glass.
• Spheres produced by the
method of G. R. Elliot et al. (Opt.
Express, 15 (2007))
• This method produces
microspheres of various sizes (20-60 µm).
• Precursor glass and spheres are
both produced at ULL.
Experimental setup
CW laser at 514 nm (Ar+ laser) as pump source resonant to the Nd3+:4I9/2→4G9/2 absorption transition. The observed emission band was that associated to the Nd3+:4F3/2→4I9/2 transition.
0 5 10 15 20
4G7/2
4G5/2, 2G7/2 2H11/2
4F9/2 4G9/2 2G9/2
Nd3+
Energy ( 103 cm-1 )
4I15/2 4I13/2
4I11/2 4I9/2
4F7/2, 4S3/2 4F5/2, 2H9/2 4F3/2
Outline
• Introduction and motivations
• Experiment details
• µ -transmission and µ -PL results
• Spectral dependence of the Q-factors
• Conclusions
Transmission on a single sphere
600 700 800 900 0
1 2 3 4 5
Intensity (arb.units)
Wavelength (nm) a)
690 700 710
b)
690 700 710
--- Outside sphere b) --- Inside sphere
• Efficient WGM excitation even if the incident beam is in free
space.
• Oscillatory-like behavior
• FSR for the oscillating signal of about 3.73 nm.
FSR for the WGM of about 3.62 nm.
• If we define the ratio among those two FSR as ΓFSR then, for this experimental case, ΓΓΓΓFSR=1.03.
R=18 µm on the CCD
A. Serpenguzel et al. J. Opt. Soc. Amer. B14 (1997) 790-795
FSR of the oscillating signal vs R
660 670 680 690 700 710 720 730
R=24.5µµµµm R=21.5µµµµm R=17µµµµm R=12µµµµm
Intensity (arb.units)
Wavelength (nm)
2 2.5 3 3.5 4
4.5 12 14 16 18 20 22 24 26
Equation y = a*x^b Adj. R-Square 0.9899
Value Standard Error
B a 0.04783 0.00756
B b -0.98583 0.05728
R (µm)
FSR (x10-3 µm)
A 1/R behavior is extracted from the fit of the FSR vs R
Clear indication of an interference
phenomenon involving
light propagating within
the microsphere
Model for explaining the origin of the oscillating–like signal
• Experimentally Γ
FSR∼ 1.03, which is close to what
given theoretically by a trajectory with p=5
Γ
FSR=1.07
2
10 cos 54º
10 cos 54º
sp
sp
m n R
FSR n R
λ
λ
=
=
µ -PL vs µ -transmission
R=9.5 µ m
Remarkable agreement in the WGM spectral positions showed by both techniques for both polarizations.
0 100 200 300 400 500 600
860 870 880 890 900 910 920
Transmission
Transmitted Intensity (arb.units)
Wavelength (nm)
TM polarisation
Photoluminescence
PL Intensity (arb.units)
TR polarisation
0 100 200 300 400 500 600
Outline
• Introduction and motivations
• Experiment details
• µ -transmission and µ -PL results
• Spectral dependence of the Q-factors
• Conclusions
Spectral dependence of the Q-factors
884.0 884.5 885.0
Q≈2300
920.5 921.0 921.5
PL Intensity (arb.units)
Wavelength (nm) Q≈5900
TM polarized resonances for a µ-sphere of R=9.5µm.
Q factors in this spectral region are dominated by absorption losses of the material with a maximum around 880 nm.
860 880 900 920 940
1 2 3 4 5 6 7
Q x103
Wavelength (nm)
1 1
1 1
1 − − − −
−
= + + +
= ∆
coupl curv
surf
mat
Q Q Q
Q Q
Q λ λ
Conclusions
• µµµµ-transmission allows significant WGM coupling efficiency even if the external beam is in free space.
• Comparison among µµµµ-PL and µµµµ-transmission demonstrated their consistency and complementarily.
• µµµµ-transmission: Provides information about the material in its
fundamental state and on the absence of local heating. Presence of an oscillating signal provide further information about the radius of the microspheres, if its material refractive index is known.
• µµµµ-PL: Provides much more intense WGM signal contribution and allows testing some of its active properties and potentialities, e.g. degree of material population inversion, lasing action, optomechanical features, etc.
