TEM images of undoped ZnO QDs and ZnO QDs doped with 2%, 5%, 10% and 20% Cd are shown in Figure A,B,C,D,E, respectively. The images reveal that the particle sizes of the Cd-doped ZnO QDs are significantly less than those of undoped ZnO QDs. Furthermore, all of the QDs are circular in shape with diameters of approximately 3 to 6
nm, and the particle size is inversely proportional to the Cd concentration. These results were consistent with the XRD studies in demonstrating that doping with Cd suppresses the growth of ZnO QD particles and that the QDs possess a hexagonal wurtzite structure with space group P63mc
, as shown in Figure .
TEM images. ZnO QDs doped with (A) 0%, (B) 2%, (C) 5%, (D) 10% and (E) 20% Cd.
XRD patterns for (a) ZnO, (b) 5% Cd-doped ZnO and (c) TOPO/ODA-modified Cd-doped ZnO QDs.
Figure shows the UV-visible absorption spectra of ZnO QDs with different concentrations of Cd. The ob-vious exciton absorption peak of ZnO QDs appears at approximately 340
nm due to the relatively large exciton binding energy of 60
meV at room temperature. For Cd-doped ZnO QDs, a blue shift was observed in the exciton absorption peak as the concentration of Cd was increased. This shift may be due to the quantum confinement effect. For the direct-band-gap semiconductor of ZnO, the band gap energy can be expressed by the following equation [3
The UV-visible absorption spectra of ZnO QDs with different concentrations of Cd. (a) 0.0%, (b)2.0%, (c) 5.0%, (d) 10.0 (e) 20.0%.
is the absorption coefficient, hν is the photon energy, A
is the edge-width parameter and Eg is the band-gap energy for direct transitions as indicated in Figure . The particle size of the ZnO QDs with different concentrations of Cd can also be estimated from the band-gap energy based on the Brus effective mass approximation formula [14
]. The diameters of these samples were calculated to be 6, 5.2, 3.6, 3.3 and 3.2
nm for pure ZnO, 2.0% Cd-doped ZnO, 5.0% Cd-doped ZnO, 10.0% Cd-doped ZnO and 20.0% Cd-doped ZnO QDs, respectively. The calculated diameters were consistent with those observed from the TEM analysis.
Plots of (αhν)2versus hν of ZnO QDs with different concentrations of Cd. (a) 0%, (b) 2%, (c) 5%, (d) 10%, (e) 20%.
Figure A illustrates that the fluorescence spectra of ZnO QDs consist of two parts: one narrow and weak peak at 385
nm is present around the ultraviolet region due to the free exciton composite, which is caused by the recombination luminescence of electron–hole pairs from the hole at the top of the valence band and the electronic states at the bottom of the conduction band [15
]; the other part is a strong and broad green emission peak at 525
nm in the visible region that arises due to the band gap of the intrinsic defect energy levels. The broad peak in the visible region is associated with structural defects, such as interstitials, oxygen vacancies and surface traps [16
]. Previous studies have shown that the properties of UV fluorescence emission peaks depend strongly on the interband transitions and the exciton recombination. Emission intensity in the ultraviolet region is increased due to the incorporation of Cd, which changed the ZnO bandgaps and promoted recombination luminescence of the exciton. The emission spectra of the ZnO QDs with different concentrations of Cd at an excitation wavelength of 365
nm are shown in Figure B. The intensity of the visible light emission peak was significantly enhanced with increasing Cd concentration. The luminescence efficiency of nanoparticles is generally believed to strongly depend on the nature of the surface because large surface-to-volume ratios cause surface defects in smaller particles [18
]. Because both Cd and Zn belong to the IIB family, they are similar in physical and chemical properties and have the same valence electron configuration: (n –
. Furthermore, RCd2+
nm) is larger than RZn2+
nm). As a result, Cd ions displace some Zn ions in the ZnO lattice as substitutional impurities. Moreover, the incorporation of Cd changes the coordination number of the cations and permits more oxygen vacancies and other defects, which may be the main reason for the enhancement of the luminescence intensity with increased Cd doping.
Emission spectra of ZnO QDs with different concentrations of Cd. (A) 318-nm and (B) 365-nm excitation. (a) 0.0%, (b) 2.0%, (c) 5.0%, (d) 10.0%, (e) 20.0%.
Figure shows the UV-visible absorption spectra for both TOPO/ODA-modified and unmodified Cd-doped ZnO QDs. The exciton absorption peaks of both samples appear at approximately 340
nm, whereas a blue shift relative to the peak position of unmodified Cd-doped ZnO QDs was observed in the exciton absorption peak for the TOPO/ODA-modified Cd-doped ZnO QDs when the TOPO/ODA mass ratio was 1:2. In the unmodified system, the QD particles easily aggregated to form clusters, and the particle growth within these clusters caused the shift. Thus, we deduced that the introduction of TOPO/ODA into the QD preparation process can effectively inhibit the aggregation of particles.
The UV-visible absorption spectra of (a) unmodified and (b) TOPO/ODA-modified Cd-doped ZnO QDs.
We sought to obtain a full understanding of the effect of these modifications on the fluorescence properties of QDs. The emission spectra, which were derived from the excitation of Cd-doped ZnO QDs with different TOPO/ODA mass ratios at 341
nm, are shown in Figure . A shift similar to that of the UV absorption peaks was observed in the emission spectra because of the quantum confinement effect and because of the increase in the band-gap width. The results indicate that the coating of the surfaces of the QDs with TOPO and Cd, combined with the coordination effects of TOPO and cadmium ions, can enhance the luminous intensity. The luminous intensity was observed to increase as the TOPO/ODA mass ratio was increased when the ratio was 1:2 or less. Excessive TOPO adhesion to the QD particles can block the absorption of the excitation light [20
], and the emission intensity decreases as a result. The results show that a 1:2 TOPO/ODA mass ratio is optimal for the generation of maximum luminous intensity.
Fluorescence spectra of Cd-doped ZnO QDs for different TOPO/ODA mass ratios. (a)1:5, (b) 1:3, (c) 1:2, (d) 1:1, (e) 2:1.
Figure shows the FT-IR spectra of both the unmodified Cd-doped ZnO QDs and the TOPO/ODA-modified QDs. The well-known stretching mode of Cd
O was observed at 1,420
. Clear Zn-O-Zn stretching modes were observed at 457
], and these stretching modes were indicative of the successful synthesis of ZnO in both cases, as previously confirmed by XRD. A peak at 1,300
was observed because of the absorption of P
O, whereas stretching modes at 2,995 and 2,923
were due to the -CH3
- groups present at the surface of the Cd-doped ZnO QDs embedded in TOPO/ODA. These results demonstrate that the polymer successfully coated the surface of the Cd-doped ZnO QDs.
FT-IR spectra of (a) unmodified and (b) TOPO/ODA-modified Cd-doped ZnO QDs.
Figure presents representative XRD patterns of the ZnO QDs (a), the 5% Cd-doped ZnO QDs (b) and the TOPO/ODA-modified Cd-doped ZnO QDs (c) synthesised via the sol–gel method. The XRD spectra in Figure a,b revealed broad peaks at 31.63°, 34.50°, 36.25°, 47.50°, 56.60°, 62.80° and 67.92°; these diffraction peaks matched the JCPDS file for ZnO (JCPDS 36–1451) and were indexed as a hexagonal wurtzite structure of ZnO with space group P63mc. Because no impurity peak associated with Cd clusters or CdO was detected, the wurtzite structure was not disturbed by the addition of small amounts of Cd2+ during the sample preparation. Therefore, the broadening of the XRD peaks (i.e., Scherrer broadening) gave a clear indication of the formation of nanosized ZnO. The particle diameters of the ZnO and the 5% Cd-doped ZnO QDs were estimated using the Debye-Scherer equation:
is the particle size, λ
is the wavelength of ra-diation used, θB
is the Bragg diffraction angle and B
is the peak width at half maximum [22
]. The XRD data, along with the TEM data presented previously, establish that doping with Cd suppresses the growth of ZnO QD particles. Furthermore, the XRD peaks the diffraction profiles of TOPO/ODA-modified Cd-doped ZnO QDs were not as sharp as in the case of the 5% Cd-doped ZnO sample, which indicated that the polymer coated the surfaces of the Cd-doped ZnO QDs. These results were also consistent with those from the FT-IR analysis.
Figure a,b shows images of (1) the ZnO QDs, (2) the 5% Cd-doped ZnO QDs and (3) the TOPO/ODA-modified Cd-doped ZnO QDs illuminated with ordinary and UV (365-nm excitation) lamps, respectively. The samples under ordinary light exhibited a pure white colour, whereas the samples under UV light showed greenish-yellow luminescence. The fluorescence properties of the synthesised ZnO, the Cd-doped ZnO and the TOPO/ODA-modified Cd-doped ZnO QDs are shown in Figure . The luminous colour is apparently due to the emission peak at approximately 500
nm in the visible region, and this broad peak is associated with structural defects, such as interstitials, oxygen vacancies and surface traps.
Fluorescence photographs of QDs under illumination by (a) an ordinary lamp and (b) UV light. (1) The ZnO QDs, (2) the 5% Cd-doped ZnO QDs, (3) the TOPO/ODA-modified Cd-doped ZnO QDs.