Rapid degradation of CdSe/ZnS colloidal quantum dots exposed to ...

APPLIED PHYSICS LETTERS 93, 173101 2008
Rapid degradation of CdSe/ZnS colloidal quantum dots exposed
to gamma irradiation
Nathan J. Withers, 1 Krishnaprasad Sankar, 1 Brian A. Akins, 1 Tosifa A. Memon, 1
Tingyi Gu, 1,a Jiangjiang Gu, 1,a Gennady A. Smolyakov, 1 Melisa R. Greenberg, 1,b
Timothy J. Boyle, 2 and Marek Osiński 1,c
1 Center for High Technology Materials, University of New Mexico, 1313 Goddard SE, Albuquerque,
New Mexico 87106-4343, USA
2 Advanced Materials Laboratory, Sandia National Laboratories, 1001 University Blvd. SE, Albuquerque,
New Mexico 87106, USA
Received 29 March 2008; accepted 2 July 2008; published online 28 October 2008
Effects of
137 Cs gamma irradiation on photoluminescent properties of CdSe/ZnS colloidal
quantum dots are reported. Optical degradation is evaluated by tracking the dependence of
photoluminescence intensity on irradiation dose. CdSe/ZnS quantum dots show poor radiation
hardness, and severely degrade after less than 20 kR exposure to 662 keV gamma photons. © 2008
American Institute of Physics. DOI: 10.1063/1.2978073
Colloidal quantum dots QDs have attracted tremendous
interest over the last few years for a large variety of applications
ranging from optoelectronic through photocatalytic to
biomedical, including applications as nanophosphors in
light emitting diodes LEDs. 1 For example, red and green
CdSe/ZnS QDs have been used as phosphors in white LEDs
based on blue-emitting InGaN-based chips. 2 CdSe/ZnS QDs
have also been suggested as scintillators for detection of alpha
particles and gamma rays. 3,4 For applications that involve
exposure of light sources to ionizing radiation, it is
important to know the levels of irradiation that would degrade
their optical properties. In this paper, we report on the
effects of gamma irradiation on photoluminescent properties
of CdSe/ZnS QDs.
CdSe cores were synthesized according to a modified
procedure of Aldana et al. 5 First, the cadmium precursor
cadmium acetate was dissolved in a coordinating solvent,
trioctylphosphine oxide TOPO. The solution was heated in
a flask under controlled argon atmosphere to 325 °C, while a
solution of selenium in trioctylphosphine TOP and toluene
was prepared in an argon-filled glovebox with 0.1 ppm of
water vapor and oxygen. When the temperature of the solution
in the flask reached 325 °C, the selenium precursor solution
was rapidly injected into the flask. When the contents
of the flask reached red color, the solution was cooled to
room temperature by removing the heat source. CdSe cores
were precipitated in a centrifuge using acetone and methanol,
and collected in hexanes. For ZnS coating, the zinc precursor
was prepared by dissolving the zinc alkyl alkoxide with oleic
acid and octadecene, while the sulfur precursor was prepared
by dissolving sulfur in octadecene. Both precursors were
heated to 200–250 °C and the zinc precursor was cooled to
60–80 °C, while the sulfur precursor was cooled to room
temperature. The CdSe nanocrystals in hexanes were mixed
with octadecene and octadecylamine in a flask and heated to
100 °C under vacuum for 30 min, and then under argon to
240 °C. At that point, three alternating injections of calculated
amounts of zinc and sulfur precursors were made at
10 min intervals. Finally the solution was cooled to room
temperature, QDs were precipitated with acetone and methanol,
and collected in hexanes.
An Eberline 1000B multiple-source gamma calibrator
was used to study the effects of irradiation on CdSe/ZnS
QDs. In order to accelerate the degradation process, the
strongest of the available sources was used, namely a 137 Cs
source with the activity of 39.7 Ci, emitting 662 keV rays.
Optical degradation was evaluated based on the measured
dependence of room-temperature RT photoluminescence
PL intensity on the irradiation dose. Using a Horiba
Jobin Yvon Fluorolog-3 spectrofluorometer, PL measurements
were performed after regular weekly periods of irradiation
to check if the QDs exhibited any signs of degradation
in their optical characteristics. In order to separate
irradiation-induced phenomena from possible effects of natural
degradation due, for example, to oxidative processes, the
irradiated sample had a control counterpart from the same
synthesis batch that was not subjected to irradiation. Assuming
that both the irradiated and control samples stored at RT
underwent the same natural aging processes and reacted to
environmental changes in the same way, we corrected the
results of PL degradation measurements of the irradiated
sample for any changes in PL intensity of the control sample
with respect to its baseline measurement performed prior to
irradiation.
The PL lifetime measurements were taken on the same
spectrofluorometer in a different configuration, allowing for
time-correlated single photon counting. Using a 375 nm
pulsed diode laser source, the resolution of lifetime measurements
was 100 ps. An integrating sphere accessory on the
spectrofluorometer was used to measure quantum efficiency
of the sample prior to and after irradiation.
Figure 1a summarizes the results of the irradiation testing.
The PL intensity was taken at the peak of PL emission,
which shifted slightly toward shorter wavelengths with increasing
dose. CdSe/ZnS QDs, initially very bright, showed
a rapid loss of light output when exposed to 662 keV rays,
and were removed from the experiment after 133.2 kR cua
On leave from: School of Electronic, Information, and Electrical Engineering,
Shanghai Jiao Tong University, China.
b Present address: CVI Laser LLC, 200 Dorado SE, Albuquerque, NM
87123.
c Electronic mail: osinski@chtm.unm.edu.
0003-6951/2008/9317/173101/3/$23.00
93, 173101-1
© 2008 American Institute of Physics
Author complimentary copy. Redistribution subject to AIP license or copyright, see http://apl.aip.org/apl/copyright.jsp

173101-2 Withers et al. Appl. Phys. Lett. 93, 173101 2008
FIG. 2. Effects of irradiation on PL lifetimes for CdSe/ZnS QDs.
FIG. 1. a Peak PL intensity in counts/second for CdSe/ZnS QDs as a
function of cumulative exposure E in kiloroentgens. b Spectral changes in
PL response of CdSe/ZnS QDs induced by 662 keV -ray irradiation.
mulative exposure. An examination of PL spectra of these
QDs revealed an irradiation-induced blueshift of the PL
emission peak from the original 579 to 570 nm Fig. 1b.
The CdSe/ZnS material system is known to be 12%
lattice-mismatched, 6 which causes the bandgap of CdSe to
reduce under strain. We tentatively explain the observed
blueshift in the PL emission peak by irradiation-induced defects
of the QD crystalline structure that lead to partial strain
relaxation.
In order to enable a comparison of radiation hardness of
CdSe/ZnS QDs with that of other materials, the exposure in
roentgens E has to be converted to an absorbed dose in rads
D using the following formula valid for a monoenergetic
source, 7
D = 0.88E en h/ CdSe / en h/ air ,
1
where h is the photon energy and en h/ X is the
mass energy-absorption coefficient for the subscripted material
X. While the energy-absorption coefficient for air at
662 keV is known to be 0.0293 cm 2 /g, 8 its value for CdSe
needs to be estimated using the constituent element data according
to the following formula: 9
en h/ CdSe = tr h/ Cd 1−f Cd g Cd − f Se g Se f Cd
+ tr h/ Se 1−f Cd g Cd − f Se g Se f Se .
2
Here, tr h/ Y Y =Cd,Se is the mass energy-transfer
coefficient, f Y is the weight fraction, and g Y is the average
fraction of secondary-electron energy that is lost in radiative
interactions. 9 Calculated from Eqs. 1 and 2, the absorbed
dose conversion factor for CdSe at 662 keV was found to be
0.899 rad/R. In terms of the absorbed dose, the CdSe/ZnS
QDs turned out to be of very poor radiation hardness, having
lost 50% of their light output after only 11.5 krad of absorbed
dose equivalent to 11.3 kradSi, rendering them
unsuitable for aerospace and terrestrial applications, where
radiation hardness up to total -ray doses of 250 Mrad is a
prerequisite. This should be contrasted with excellent radiation
hardness of GaN/InGaN multiple-quantum-well LEDs,
which can sustain 250 MradSi from 60 Co source with
45% loss in their PL intensity. 10 It should be pointed out,
however, that CdSe/ZnS QDs still outperform the standard
NaI:Tl scintillating crystals, which according to Normand et
al. 11 rapidly lose the light output after exposure to merely
500 rad of absorbed dose from 60 Co source.
Figure 2 shows PL decay data for the CdSe/ZnS QDs,
obtained prior to irradiation and after 120 krad of absorbed
dose. The observed changes in PL lifetime were accompanied
by a reduction in quantum yield from 23.4% to 0.2%. A
triple-exponential fit was used to achieve good agreement
with each of the measured PL decay curves in Fig. 2. The
decay time constants and the relative percentage weights of
the decay components are summarized in Table I.
The current understanding of the exciton decay dynamics
in colloidal nanocrystals is still very fragmentary, even
for the most thoroughly investigated CdSe QDs. Initial studies
of the band-edge luminescence lifetimes revealed that the
low-temperature exciton lifetime of CdSe QDs was orders of
magnitude longer microseconds versus 1 ns than that in
the bulk CdSe. 12,13 These observations were initially ascribed
to surface localization of the photogenerated hole. 12,13 However,
further experiments using fluorescence line-narrowing
spectroscopy on surface-modified CdSe and CdSe/ZnS
QDs 14 combined with theoretical analysis 15 established that
the energetics and dynamics of the excitonic PL could be
understood in terms of the intrinsic band-edge exciton fine
structure, with surface effects being responsible solely for
the nonradiative relaxation pathways. Temperature and sizedependence
of radiative exciton lifetimes and, in particular,
TABLE I. Results of a multiexponential fit to the data shown in Fig. 2.
Sample A 1 1 ns A 2 2 ns A 3 3 ns
Control 6.31 1.3 67.46 16.24 26.23 41.5
Irradiated 14.78 0.73 39.17 9.08 46.05 32.6
Author complimentary copy. Redistribution subject to AIP license or copyright, see http://apl.aip.org/apl/copyright.jsp

173101-3 Withers et al. Appl. Phys. Lett. 93, 173101 2008
the anomalously long low-temperature exciton lifetimes are
well explained by thermal mixing of “dark” and “bright”
exciton states separated by the size-dependent energy
gap. 16–18
The dynamics of band-edge PL decay is typically multiexponential
at short times, while at longer times, after the
excitons have thermalized, the radiative decay closely follows
a single exponential. It is this, longer, singleexponential
decay component, demonstrating strong temperature
and size dependence and largely insensitive to
surface nonradiative effects, that is identified as exciton radiative
lifetime. 16 The faster decay dynamics at short times,
on the other hand, is clearly affected by nonradiative defectrelated
pathways. Previous work has shown that the quality
of the quantum dot synthesis determines the shape of the
time-resolved emission decay. 19 As the surface and lattice
structure of CdSe core is improved and the quantum yield
increases to as high as 80%, the faster multiexponential dynamics
at short times effectively disappears. The decay becomes
single exponential in character, with the lifetime of
30 ns.
We have observed the drop in quantum yield from
23.4% to 0.2% after 120 krad of absorbed dose, which
results from gamma-irradiation-induced nonradiative
recombination channels and, within the framework of the
“dark” exciton model, is expected to affect the short
lifetime component in the PL decay data of Fig. 2. Using
the fastest decay component from Table I, and assuming
that it originates from both radiative and nonradiative
decay of the “bright” exciton state, we can estimate the
irradiation-induced nonradiative exciton lifetime as follows:
Prior to irradiation, 1/ 1 =1/ r +1/ nr , while after irradiation
1/ * 1
=1/ r +1/ nr +1/ nr =1/ 1 +1/ nr , where
nr is the
irradiation-induced nonradiative exciton lifetime. With 1
=1.3 ns and * 1
=0.73 ns, we arrive at nr 1.66 ns. We note,
however, that this mechanism alone cannot account for the
dramatic reduction in the quantum yield. Indeed, if the
nonradiative decay of the bright exciton were the only nonradiative
recombination pathway, the value of * 1
would have
to be 65 times shorter than that given in Table I. We explain
the observed quantum yield drop by a very strong
irradiation-induced nonradiative pathway that involves direct
trapping of excited carriers by irradiation-induced defects,
without formation of the excitons.
In conclusion, CdSe/ZnS colloidal nanocrystals have
been tested for the effects of 662 keV irradiation from a
137 Cs source, revealing poor radiation hardness for that QD
material. Irradiation-induced reduction in photoluminescence
lifetime was observed, accompanied by a dramatic deterioration
of PL quantum efficiency. We ascribe the shortening of
PL lifetime to irradiation-enhanced pathway for nonradiative
recombination of bright exciton states, while the dramatic
drop of quantum efficiency is explained by nonradiative processes
that do not involve excitonic states, namely, by direct
trapping of excited carriers to deep energy levels associated
with defects created by the irradiation.
This work was supported by the NSF Grant Nos. IIS-
0610201, CBET-0736241, and DGE-0549500. The authors
express their gratitude to members of the UNM Department
of Safety and Risk Services: Jim De Zetter, Marybeth
Marcinkovich, Ralph M. Becker, Marj Walters, and Tom
Rolland for their regular assistance with the use of Eberline
1000B multisource gamma calibrator. Dr. Markku Koskello
and Don Jacoby of Canberra Albuquerque, Inc., are gratefully
acknowledged for the loan of a calibrated Canberra
Radiac Meter Geiger-Müller counter, which was used to calibrate
the Eberline 137 Cs source used in irradiation tests.
1 Y. Q. Li, A. Rizzo, R. Cingolani, and G. Gigli, Microchim. Acta 159, 207
2007.
2 H.-S. Chen, C.-K. Hsu, and H.-Y. Hong, IEEE Photonics Technol. Lett.
18, 193 2006.
3 S. E. Létant and T.-F. Wang, Appl. Phys. Lett. 88, 103110 2006.
4 S. E. Létant and T.-F. Wang, Nano Lett. 6, 2877 2006.
5 J. Aldana, Y. A. Wang, and X.-G. Peng, J. Am. Chem. Soc. 123, 8844
2001.
6 R.-G. Xie, U. Kolb, J.-X. Li, T. Basché, and A. Mews, J. Am. Chem. Soc.
127, 7480 2005.
7 P. W. Cattaneo, IEEE Trans. Nucl. Sci. 38, 894 1991.
8 K. Eckerman, RADIOLOGICAL TOOLBOX COMPUTER PROGRAM, Oak Ridge National
Lab., 2003.
9 F. H. Attix, Introduction to Radiological Physics and Radiation Dosimetry
Wiley, New York, 1986, p. 155.
10 R. Khanna, S. Y. Han, S. J. Pearton, D. Schoenfeld, W. V. Schoenfeld, and
F. Ren, Appl. Phys. Lett. 87, 212107 2005.
11 S. Normand, A. Iltis, F. Bernard, T. Domenech, and P. Delacour, Nucl.
Instrum. Methods Phys. Res. A 572, 7542007.
12 M. G. Bawendi, P. J. Carroll, W. L. Wilson, and L. E. Brus, J. Chem. Phys.
96, 946 1992.
13 M. Nirmal, C. B. Murray, and M. G. Bawendi, Phys. Rev. B 50, 2293
1994.
14 M. Kuno, J. K. Lee, B. O. Dabbousi, F. V. Mikulec, and M. G. Bawendi,
J. Chem. Phys. 106, 9869 1997.
15 A. L. Efros, M. Rosen, M. Kuno, M. Nirmal, D. J. Norris, and M. Bawendi,
Phys. Rev. B 54, 4843 1996.
16 S. A. Crooker, T. Barrick, J. A. Hollingsworth, and V. I. Klimov, Appl.
Phys. Lett. 82, 2793 2003.
17 O. Labeau, P. Tamarat, and B. Lounis, Phys. Rev. Lett. 90, 257404 2003.
18 C. D. Donega, M. Bode, and A. Meijerink, Phys. Rev. B 74, 085320
2006.
19 C. D. Donega, S. G. Hickey, S. F. Wuister, D. Vanmaekelbergh, and A.
Meijerink, J. Phys. Chem. B 107, 489 2003.
Author complimentary copy. Redistribution subject to AIP license or copyright, see http://apl.aip.org/apl/copyright.jsp