articol eficienta emisiei prin upconversie

8/19/2019 Articol Eficienta Emisiei Prin Upconversie

1/3


2/3

calculated the corresponding Judd–Ofelt parameters. In this

letter, we extend the study by investigating the upconversion

emission properties after excitation with NIR radiation 980

nm.

Figure 1 presents the emission spectra of the

ZnO–TeO2 – Er2O3 glass upon direct excitation with 488 nm

and after exciting into the 4 I 11/2 (exc980 nm) excited state

of the erbium ion. Green emission is observed and is attrib-

uted to the transitions from the thermalized 2 H 11/2 and 4S 3/2

states to the 4 I 15/2 ground state, centered at 535 and 550 nm,

respectively. Red and NIR emission was observed from the4F 9/2 670 nm,

4 I 9/2 800 nm excited states to the

ground state and for the 4S 3/2→4 I 13/2 transition at 850 nm.Following 980 nm excitation, two weak bands were detected

at 455 and 490 nm. They are assigned to the 4F 5/2→

4 I 15/2 and 4F 7/2→

4 I 15/2 transitions, respectively. It should

be noted that the peak shape of the ( 2 H 11/2 , 4S 3/2)→

4 I 15/2and 4F 9/2→

4 I 15/2 transitions in the upconversion spectrum

differs from those in the direct emission spectrum. We have

also performed the upconversion at 971, 978, and 986 nm

and have observed that the above-mentioned transitions have

slightly different peak shapes in each upconversion spectrum

depending on the excitation wavelength. We hypothesize that

each exciting wavelength tunes into a specific ESA process

specific to only a subset of Er3 sites in the glass matrix. An

exhaustive site-selective upconversion study will elucidate

this point.

The upconverted green luminescence is very bright and

can be observed by the naked eye after pumping with as low

as 20 mW. The dependence of the upconverted luminescence

on pump power is quadratic, thereby indicating a two-photon

process. This two-step upconversion can occur by only two

distinct processes capable of populating the green emitting

( 4S 3/2) level; energy transfer ET and excited-state absorp-

tion. No inflection was observed in the power study, thereby

eliminating photon avalanche PA as a possible

mechanism.8

The temporal behavior of the green upconverted( 2 H 11/2 ,

4S 3/2)→4 I 15/2 emission is shown in Fig. 2. The rise

time, following 980 nm excitation, is 440 s, which is in

good agreement with the lifetime of the 4 I 11/2 level 400 smeasured by Sidebottom et al. at 300 K.9 Clearly, the long-

lifetime state ( 4 I 11/2) contributes as an intermediate state in

the upconversion process. Examination of the energy-level

diagram for Er3 shows that there is an energy level lying at

twice the excitation energy of 980 nm. So, we postulate that

the upconversion occurs via an ESA process Fig. 3. The

Er3 ion is excited to the 4 I 11/2 level, after which a second

photon from the pump beam brings the ion to the 4F 7/2 level.

Nonradiative relaxation in turn populates the 4S 3/2 level. In

FIG. 1. Room-temperature luminescence of 19ZnO– 80TeO2 –1Er2O3 upon

excitation with a 488 nm and b 980 nm. i 2 H 11/2→4 I 15/2 , ii

4S 3/2→

4 I 15/2 , iii 4F 9/2→

4 I 15/2 , iv 4 I 9/2→

4 I 15/2 , and v 4S 3/2→

4 I 13/2 .

FIG. 2. Temporal behavior of the 19ZnO– 80TeO2 –1Er2O3 upconverted

emission following 980 nm excitation.

FIG. 3. Energy-level diagram of Er3 ions in 19ZnO– 80TeO2 –1Er2O3

showing the two-photon ESA and PET upconversion processes responsiblefor populating the 4F 7/2 and

4F 5/2 levels, respectively (exc980 nm).

1753Appl. Phys. Lett., Vol. 80, No. 10, 11 March 2002 Vetrone et al.

Downloaded 06 Mar 2002 to 132.205.24.80. Redistribution subject to AIP license or copyright, see http://ojps.aip.org/aplo/aplcr.jsp


3/3

general, it has been shown that energy transfer between dop-

ant ions can be neglected in samples with a concentration of

0.5 mol % or lower.10 The sample used in this study is doped

with 1 mol % Er3 and as a result, ET should be taken into

account in the upconversion process. Inspection of the blue

region of the upconversion spectrum gives further proof of

this. We observe emission from the 4F 5/2 level, however, two

photons of 980 nm do not have the necessary energy to

populate this level. A nonradiative decay to this level from athree-photon ESA process is ruled out. The energy-level dia-

gram reveals that there is no energy level resonant at three

times the excitation wavelength (310 200 cm1). Simi-

larly, if the ion nonradiatively decays to the 4 I 13/2 , an excit-

ing photon has no resonant energy level for which to popu-

late. Therefore, an energy transfer upconversion mechanism

must be operative. This is evident as the decay time of the

green (4S 3/2→4 I 15/2) upconverted emission is longer than ex-

citing directly in the emitting level. The lifetime of the

( 2 H 11/2 ,4S 3/2) states at 488 nm is 22 s, while upon excita-

tion at 980 nm the lifetime of the same states is 40 s. The

ET upconversion Fig. 3 occurs by the following mecha-

nism:

4 I 11/2 , 4 I 11/2→

4F 5/2 , 4 I 15/2.

Energy transfer upconversion involves another Er3 ion in

close proximity. After the ion is excited to the 4 I 11/2 level by

the laser beam, a neighboring Er3 ion also in the 4 I 11/2 state

transfers its energy to the initial ion, thereby exciting it to the4F 5/2 level. We point out that there is not perfect resonance

between the 4 I 11/2 and 4 F 5/2 states. In order for the

4F 5/2 state

to be populated after 980 nm excitation, an energy gap of

1750 cm1 must be satisfied. The highest phonon energy in

the zinc tellurite host11 is 773 cm1, therefore, two or three

phonons would fill the necessary gap.

The 19ZnO–80TeO2 –1Er2O3 glass yields a predomi-

nantly green emission ( 2 H 11/2 ,4S 3/2)→

4 I 15/2)] in the upcon-

version spectrum excited by 980 nm radiation. The upcon-

version efficiency for this transition was, therefore,

calculated. Upconversion efficiency is given by4

P emittedP absorbed

emited green light powerabsorbed IR light power

. 1

The emitted green light power after NIR excitation can be

estimated by comparing it with that obtained after direct 488

nm excitation. It is necessary that we first determine lumi-

nescence quantum yield q for 488 nm excitation:12

qemitted light power

absorbed radiation power

exp

R, 2

where exp is the lifetime of the level determined experimen-

tally after excitation at 488 nm while R is the radiative

lifetime. The radiative lifetime of the upper level i is cal-

culated by13

R1

j A i j, 3

where j is the summation to all the lower levels. The total

spontaneous emission probabilities, A i j , were calculated in a

previous paper.7 The radiative lifetime of the thermalized

(2 H 11/2 ,4S 3/2) states was obtained from the Judd–Ofelt

analysis and was determined to be 186 s for the ZnO–TeO2glass. The experimental lifetime at room temperature, fol-

lowing 488 nm excitation, was found to be 22 s. The lumi-

nescence quantum yield was calculated to be 12%. The emis-

sion light intensity can be determined as follows:

Pemitted qabsorbed radiation density. 4

Also, the absorbed light intensity, Pabs , can be calculatedfrom the absorption coefficient, the incident light power, and

the sample length. It should be noted that using this method

requires calibration of the detecting equipment. In order to

avoid the calibration step, we incorporate the ratio of the

upconversion luminescence and direct excitation lumines-

cence intensities4 obtained under identical collecting condi-

tions:

q P absvisibleP absIR I emittedupconversion

I emitteddirect , 5

where P abs visible is the absorbed light power for direct

excitation with 488 nm and P abs IR is the absorbed light

power for upconversion with 980 nm excitation. The upcon-version efficiency was measured using a pump power of 350

mW 880 W/cm2. We obtained an upconversion efficiency

of 0.16%

In this letter, we have reported and discussed the visible

upconversion emission of a 19ZnO–80TeO2 –1Er2O3 glass

following excitation into the 4 I 11/2 level, which yields a vi-

sually dominant green emission, as well as having evaluated

the upconversion efficiency 0.16% for the emission from

the thermalized ( 2 H 11/2 , 4S 3/2) states. We have shown that

the upconversion occurs via a two-photon ESA and PET pro-

cess, and the temporal studies revealed that 4 I 11/2 is the in-

termediate level from which the upconversion occurs.

The authors gratefully thank Erica Viviani Università di

Verona, Italy for expert technical assistance. The authors

acknowledge the Natural Science and Engineering Research

Council of Canada and MURST of Italy, for financial sup-

port.

1 W. M. Yen, in Optical Spectroscopy of Glasses, edited by I. Zschokke

Reidel, Dordrecht, 1986, pp. 23–64.2 W. Ryba-Romanowski, S. Golab, and G. Dominiak-Dzik, J. Phys. Chem.

Solids 54, 153 1993.3 C. B. Layne, W. H. Lowdermilk, and M. J. Weber, Phys. Rev. B 16, 10

1977.4 Z. Pan, S. H. Morgan, K. Dyer, A. Ueda, and H. Liu, J. Appl. Phys. 79,

8906 1996.5 N. Jaba, A. Kanoun, H. Mejri, A. Selmi, S. Alaya, and H. Maaref, J. Phys.:

Condens. Matter 12, 4523 2000.6 A. Yariv, Optical Electronics in Modern Communications, 5th ed. Oxford

University Press, Oxford, U.K., 1997.7 R. Rolli, K. Gatterer, M. Wachtler, M. Bettinelli, A. Speghini, and D. Ajò,

Spectrochim. Acta, Part A 57, 2009 2001.8 M. F. Joubert, Opt. Mater. 11, 181 1999.9 D. L. Sidebottom, M. A. Hruschka, B. G. Potter, and R. K. Brow, J.

Non-Cryst. Solids 222, 282 1997.10 J. P. van der Ziel, F. W. Ostermayer, and L. G. Van Uitert, Phys. Rev. B 2,

4432 1970.11 T. Sekiya, N. Mochida, and A. Ohtsuka, J. Non-Cryst. Solids 168, 106

1994.12 D. C. Yeh, W. A. Sibley, M. Suscavage, and M. G. Drexhage, J. Appl.

Phys. 62, 266 1987.13 M. J. Weber, Phys. Rev. 157, 262 1967.

1754 Appl. Phys. Lett., Vol. 80, No. 10, 11 March 2002 Vetrone et al.

Downloaded 06 Mar 2002 to 132.205.24.80. Redistribution subject to AIP license or copyright, see http://ojps.aip.org/aplo/aplcr.jsp

articol eficienta emisiei prin upconversie

Documents