Surface Science 601 (2007) 2667–2670
www.elsevier.com/locate/susc
Synthesis and characterization of brightly photoluminescent
CdTe nanocrystals
J. Kolny-Olesiak b, V. Kloper a, R. Osovsky a, A. Sashchiuk a, E. Lifshitz
b
a,*
a
Department of Chemistry and Solid State Institute, Technion, Haifa, Israel
Energy and Semiconductor Research Laboratory, Department of Physics, University of Oldenburg, Germany
Available online 4 January 2007
Abstract
A new synthesis procedure for the preparation of spherical shaped CdTe nanocrystals (NCs) is presented, exhibiting bright luminescence with exceptionally high quantum efficiency (up to 85%). The growth of these NCs occurs in a non-coordinating solvent, octadecene, with the addition of oleic acid/tri-octylphosphine stabilizers, CdO as a precursor for the Cd monomers and additional Cd
metal particles as a supplementary Cd reservoir source. The dependence of the crystalline quality and the optical properties of the CdTe
NCs, on the initial Cd:Te precursors’ molar ratio, and the reaction duration were investigated. It was demonstrated that the NCs’ properties improved significantly as the initial Cd:Te molar ratios are increased. The obtained NCs’ properties were correlated with measurements of the Cd0 concentration in Cd metal particles, CdTe NCs and in Cd monomer solutions.
2006 Elsevier B.V. All rights reserved.
Keywords: Colloidal; CdTe; Nanocrystals; PL; Quantum efficiency
1. Introduction
In the last two decades a considerable progress has been
made in the fundamental [1] and applied physical research
of semiconductor nanocrystals (NCs) [2,3]. Colloidal CdTe
NCs, in particular, show an increasing interest due to their
large Bohr radius (7.3 nm) and the relatively small band
gap (1.475 eV), leading to a pronounced quantum size effect in a range of CdTe NCs with a diameter up to
10 nm. In addition, the absorption and emission spectra
of these NCs can reach the near-IR spectral regime. The
synthesis of high-quality NCs requires specific conditions
of monomer reactivity and stabilizing ligands in order to
control the monodispersity and optical properties of the
growing NCs [4].
In this paper we present a new synthesis procedure,
using two sequential stages for the preparation of highquality CdTe NCs with photoluminescence (PL) quantum
*
Corresponding author. Tel.: +972 48293987.
E-mail address: ssefrat@tx.technion.ac.il (E. Lifshitz).
0039-6028/$ - see front matter 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.susc.2006.12.013
efficiency (QE) up to 85%. Using crystalline Cd particles,
formed in situ in the first stage as an additional source
for Cd monomers, the growth of CdTe NCs occurred in
a non-coordinating solvent, octadecene (ODE), during
the second stage, with OA/TOP stabilizers.
2. Experimental
A Te precursor solution was prepared by dissolving Te
(0.0128 g) in tri octylphosphine (TOP, 0.2112 g), and octadecene (ODE) to a total amount of 2 mL. The Cd precursor solution was prepared by mixing CdO (0.0256 g) and an
oleic acid (OA, 200 lL)/ODE(10 mL) mixture. When
heated under Ar flow at 300 C all CdO was dissolved in
the OA/ODE mixture. Further heating to 310 C, led to
a formation of a grey precipitate, which was characterized
as crystalline Cd particles, and considered as the first stage
of the reaction. The second stage of the reaction included
an instantaneous injection of the Te precursor solution into
the Cd solution at 310 C. The morphology and size of the
Cd precipitate and CdTe NCs were determined by high
2668
J. Kolny-Olesiak et al. / Surface Science 601 (2007) 2667–2670
resolution scanning electron microscopy (HR-SEM) and
transmission electron microscopy (HR-TEM). The optical
properties of CdTe NCs were studied using absorption
and photoluminescence measurements. The amount of
Cd atoms in each fraction was determined by an atomic
absorption spectroscopy (AAS). The PL QE was measured
by comparing the PL intensity with that of a commercial
Rhodamine 6G with a known QE of 95%. All the measurements were carried out at room temperature.
3. Results and discussion
Fig. 1A presents a HR-SEM image that reveals that the
grey precipitate, formed during the first stage of the reaction, consists of Cd particles with average size of
100 nm and with well pronounced facets. However, after
a Te injection, the facets of the Cd particles disappear
(Fig. 1B), suggesting the occurrence of a chemical reaction
and a return of Cd monomers to the reaction solution.
Fig. 1C shows a HR-TEM image of CdTe NCs with a
diameter of 4.0 nm, indicating the formation of spherical
NCs with a 5% size distribution with a cubic zinc blende
structure (not shown). Fig. 1D shows the absorption spectra (dashed line) and the PL spectra (solid line) of various
aliquots of the CdTe NCs. The lower exciton energy in the
absorption spectra varies between 2.03 eV and 2.32 eV,
corresponding to NCs’ sizes between 3.1 nm and 3.8 nm.
These 1S-exciton bands have a full width at half maximum
(FWHM) of 100 meV. The PL spectra of the 3.1 nm–
3.8 nm samples are Stokes shifted with respect to the corresponding 1S exciton absorption band by 75 meV and
36 meV respectively.
In order to find the optimal conditions for the fabrication of high quality CdTe NCs, Cd atom concentrations
and absorption data versus reaction time were investigated
as a function of initial Cd:Te molar ratios. Fig. 2A shows
the percentage of Cd atoms that were lost from the Cd precursor solution to form the Cd particles. The half bold triangles (Fig. 2A) represent the percentage of Cd prior to an
injection of the TOP:Te solution.
The bold and open triangles represent the percentage of
Cd atoms after an injection of TOP:Te, when the initial
molar ratios are of Cd:Te = 1.25:1 and Cd:Te = 1:1.25,
respectively. Fig. 2A shows that the percentage of Cd
atoms in Cd precipitate increases gradually as long as the
TOP:Te is avoided. However, after the Te injection, the
growth of the Cd precipitate is stopped. The suppression
of further Cd precipitation after the Te injection, suggests
the occurrence of a Cd0 M Cd2+ equilibrium, which regulates the free Cd monomer concentration in the solution,
meanwhile permitting a slower and controlled growth of
spherical CdTe NCs, size focusing, excellent crystallinity
and improved surface properties. The plots in Fig. 2A also
show that the component of Cd monomers that form the
precipitate of Cd particles does not influence the growth
mechanism of CdTe NCs by different initial Cd:Te molar
B
A
1 m
D
C
PL Intensity (a.u.)
3.8 nm
8nm
Absorption (a.u.)
2 0 0 nm
3. 5 nm
3.1 nm
1.6
2.0
2.4
2.8
3.2
Energy (eV)
Fig. 1. HR-SEM image of Cd0 particles before (A) and after (B) the TOP:Te precursor injection. HR-TEM image of CdTe NCs (C). Absorption (dashed
line) and PL (solid line) spectra of CdTe NCs with average diameters between 3.1 and 3.8 nm, all recorded at room temperature (D).
2669
J. Kolny-Olesiak et al. / Surface Science 601 (2007) 2667–2670
A
1:1.25
100
B
90
90
80
80
70
70
20
20
10
10
1.25:1
C
20
0
[Cd /CdO] %
30
100
10
0
0
0 50 100 150 200
Time (sec)
0
0 20 40 60 80 100
Time (sec)
0 50 100 150 200
Time (sec)
Fig. 2. Plots of (Cd0/CdO) · 100 versus the reaction time: (A) in Cd0
precipitate before ( ) and after (m,n) the Te injection. The (m,n)
designate two synthesis with different Cd:Te ratios, as indicated in the text;
(B) in CdTe NCs (s) and in monomers (h) for a synthesis with initial
Cd:Te molar ratio of 1:1.25; and (C) in CdTe NCs (d) and in monomers
(j) for a synthesis with initial Cd:Te molar ratio of 1.25:1; The arrows
designate the time of Te precursor injection.
ratios. Fig. 2B and C show Cd atom concentration of
monomers (squares) and those that formed the CdTe
NCs (circles) for two different initial Cd:Te molar ratios.
Fig. 2B and C show Cd atom concentration for samples
with an initial Cd:Te molar ratio of 1:1.25 (Fig. 2B) and
1.25:1 (Fig. 2C).
The plots in Fig. 2B and C show that in case of Cd excess only 75% of Cd monomers form the CdTe NCs
NCs diameter (nm)
A
(Fig. 2C, bold circles), while in Te excess 90% of Cd monomers react to form the CdTe NCs (open circles in Fig. 2B).
Fig. 3 represents plots of calculated [5] average diameter
(3A), concentration (3B), absorption FWHM (3C) and PL
QE (3D) of various CdTe NCs, versus the reaction time.
The aliquots were taken from reactions with different initial
Cd:Te molar ratios as indexed in the inset of Fig. 3A. It
should be noted that the preliminary CdO concentration
was always constant.
Fig. 3A shows that in case of excess Cd (bold symbols),
the NCs grow to a certain size and then stop growing while
their concentration drops (Fig. 3B). Fig. 3C shows also a
constant size distribution broadening indicated by the
FWHMs of the absorption peaks. This size distribution
broadening can indicate that the CdTe NCs growth can
be characterized by the Oswald ripening stage in which
the larger particles continue to grow and the smaller ones
get smaller and finally dissolve. There is a shortage of Te:
due to the CdTe NCs growth, the Te is finally depleted.
Therefore, as shown in Fig. 2C, only 75% of Cd monomers are used to form the CdTe NCs, and 20% of monomers stay unrelated in solution. In the case of excess Te
(open symbols in Fig. 3A) fast NCs growth is observed,
while their concentration is maintained constantly over
time (Fig. 3B), and the NCs’ size distribution is focused
(Fig. 3C). These NCs’ growth properties are correlated
with results shown in Fig. 2B and suggest that about 90%
of Cd monomers react to form the CdTe NCs while only
10% of monomers stay in solution. Even though the NCs
B
3. 8
3. 6
Cd :Te
3. 4
2: 1
1. 5: 1
1. 25 :1
1: 1
1: 1.25
1: 1.5
1: 2
3. 2
3. 0
0
50
100
150
NCs concentration (a.u.)
40
0
200
50
Time (sec)
40
D 100
38
80
PL QE%
Absorption FWHM (nm)
C
36
34
32
28
0
100
150
Time (sec)
200
40
20
50
150
60
30
0
100
Time (sec)
200
0
50
1 00
150
200
Time (sec)
Fig. 3. CdTe NCs diameter (A); the concentration of each aliquot (B); the FWHM of the NCs 1st absorption band (C); and PL QE (D) of CdTe NCs,
prepared with various initial Cd:Te molar ratios as indicated in the inset of (A).
2670
J. Kolny-Olesiak et al. / Surface Science 601 (2007) 2667–2670
produced with an excess Te reaction have better size distribution and conservation of NCs concentration during the
reaction time, the NCs produced in conditions with an excess of Cd grow more slowly, improving their surface passivation, which leads to improved optical properties (PL
QE) as can be seen in Fig. 3D. The plots in Fig. 3D reveal
that as the initial Cd:Te molar ratios are higher, the QE improved significantly and reaches 85%.
4. Conclusions
This paper presented the occurrence of in situ precipitation of Cd particles prior to the formation of CdTe NCs
and the influence of this precipitation on the NCs growth.
The experimental evidences shown above suggest that the
existence of Cd0 precipitates induces a Cd0 M Cd2+ equilibrium with the reaction solution, leading to a regulation of
the Cd supply during the NCs growth, revealing the formation of a uniform spherical shape CdTe NCs. The spectroscopic evidence; narrow absorption, emission lines, and the
high emission QE revealed the formation of high quality
samples resulting from the initial excess of Cd.
Acknowledgments
This project was supported by the German Israel Foundation (GIF) contact no. #156/03-12.6 and by the
German-Israel Program (DIP) project no. #D 3.2 and Niederzaksen foundation project. Dr. A. Sashchiuk expresses
her deep gratitude to the Ministry of Absorption of the
State of Israel for the Giladi Fellowship.
References
[1] C. Burda, X. Chen, R. Narayanan, M.A. El-Sayed, Chem. Rev. 105
(2005) 1025.
[2] A.J. Nozik, Physica E 14 (2002) 115.
[3] A. Shavel, N. Gaponik, A. Eychmuller, Chem. Phys. Chem. 6 (2005)
449.
[4] W.W. Yu, Y.A. Wang, X. Peng, Chem. Mater. 15 (2003) 4300.
[5] W.W. Yu, L. Qu, W. Guo, X. Peng, Chem. Mater. 15 (2003) 2854.