Journal of Nanoparticle Research

, Volume 12, Issue 1, pp 101–109

One-pot synthesis of oleic acid-capped cadmium chalcogenides (CdE: E = Se, Te) nano-crystals


    • Nanoscience LaboratoryCentre for Materials for Electronics Technology (C-MET)
  • K. Srinivasa Rao
    • Nanoscience LaboratoryCentre for Materials for Electronics Technology (C-MET)
  • K. R. Patil
    • National Chemical Laboratory
  • V. N. Singh
    • Thin Film Laboratory, Department of PhysicsIndian Institute of Technology (IIT)
  • B. R. Mehta
    • Thin Film Laboratory, Department of PhysicsIndian Institute of Technology (IIT)
Research Paper

DOI: 10.1007/s11051-008-9581-y

Cite this article as:
Khanna, P.K., Srinivasa Rao, K., Patil, K.R. et al. J Nanopart Res (2010) 12: 101. doi:10.1007/s11051-008-9581-y


Surface-capped CdSe and CdTe nano-crystals (NCs) have been synthesized using cadmium acetate, oleic acid and respective tri-octylphosphine chalcogenide (TOPE; E = Se/Te) in diphenyl ether (DPE). Well-dispersed CdSe particles showed two absorption bands at the region of 431–34 and 458–60 nm in optical absorption study. A band-edge emission resulted at 515 nm with an excitation energy of 400 nm, in its photoluminescence (PL) spectrum. Similarly, UV–visible absorption study of CdTe revealed an absorption band at <700 nm. The broadened X-ray diffraction (XRD) pattern showed that at higher reaction temperature cubic CdSe but hexagonal CdTe can be obtained with crystallite size of <10 nm. Scanning electron microscopy (SEM) and transmission electron microscopy (TEM) revealed that agglomerated particles are of spherical nature. The inter-planar spacing in CdTe was measured to be 0.406 nm, a characteristic of (100) lattice plane in hexagonal CdTe. X-ray photoelectron spectroscopy (XPS) showed that CdSe NCs have better air stability stable than CdTe. Presence of organic moiety around the semiconductor particles was confirmed by infra-red (IR) spectroscopy.


SemiconductorsSurface cappingChemical synthesisPhotoluminescence


The quantum dots (QDs) are the tiny particles broadly with particle size below 20 nm, as the optimum size quantization can be attained below this range. It is reported that optimum size variation in CdSe nano-crystals (NCs) due to quantum confinement is from 01 to 11 nm that means it contain 10–10,000 atoms. As the size decreases, number of atoms on the surface increases, e.g. an absorption at about 400 nm for crystal size of 1.0–1.5 nm may have about 80–90% atoms on the surface. Therefore, less number of atoms on the surface may result loss of optical properties and shift in absorption frequency to red region of the visible spectrum due to increase in particle size (Yen et al. 2003; Rosenthal et al. 2007; Colvin et al. 1994).

Further, it is well-known that band gap of semiconductor particles can be tuned by varying the particle size and surface morphology; thus, better optical properties can be achieved with decreasing particle size of the QDs. The energy gap between the valence and conduction band can be thus manipulated by variation in the particle size of the QDs. There are several area of applications for high-quality NCs of semiconductors, e.g. in luminescent devices, as biological markers, in lasers, light emitting diodes and in telecommunication including nano-electronics, photonics, photovoltaics, etc. (Yen et al. 2003; Rosenthal et al. 2007; Raevskaya et al. 2006; Sashchiuk et al. 2004; Colvin et al. 1994; Alivisatos 1998). Quantum dots emitting range of colours into the entire visible spectrum have been reported by researchers, but the red-light emitting CdSe NCs are perhaps the most straight forward when it comes to their synthesis (Wang and Seo 2006).

One of the biggest challenges is to isolate free standing QDs that show light emission even after long storage. Normally organic or polymeric surfactants are utilized to synthesized quantum dots of excellent quality. Quantum dots can be isolated from the preparation vessel via single-stage or multi-staged centrifugation through size-selective precipitation (generally through combination of lower alcohols and non-solvent such as hexane, toluene, etc.). By adopting advanced synthetic methodology, one can prepare QDs of hydrophilic and hydrophobic nature thereby extending the scope of their applications. Amongst the binary semiconductors, CdSe and CdTe QDs are best candidates for tuned band-gap energy in the entire visible region (Lee et al. 2007; Han et al. 2006; Firth et al. 2004; Liu et al. 2008; Diao et al. 2007). Recent advances have shown that CdTe QDs are excellent candidates in probing applications, e.g. determination of cationic surfactants (CS), specifically cetyltrimethylammonium bromide (CTAB), dodecyltrimethylammonium bromide (DTAB) and cetylpyridinium chloride (CPC), via a fluorescence quenching method which employed water-soluble luminescent CdTe QDs modified with thioglycolic acid (TGA). Large number of surfactants have been reported for preparing QDs of excellent quality that include long carbon chain carboxylic acids (e.g. oleic acid, myristic acid and stearic acid), alkyl amines and trioctylphosphine/trioctylphosphine oxide (TOP/TOPO) (Qu et al. 2004; Nie et al. 2004; Murray et al. 1993, 2001). There are a number of synthetic strategies reported in the literature including simple beaker chemistry to advanced organometallic chemistry as well ultrasonic techniques for synthesis of CdSe and CdTe in aqueous as well as organic medium (Murray et al. 1993, 2001; Chen and Gao 2002; Deng et al. 2003; Peng and Peng 2001; Afzaal et al. 2003; Trindale et al. 1999; Chu and Liu 2006; Khanna et al. 2004a, b, 2008 unpublished results).

Present study is based on our chemistry expertise where a range of reactions can be adopted and be further simplified to make QD synthesis more attractive and reachable to common researchers and users. In this article, we have directed efforts to isolate surface-capped CdE nano-particles via use of oleic acid and elemental chalcogen in TOP/diphenyl ether. Use of TOPE (E = Se or Te) by normal dissolution of the chalcogen in TOP at less than 100 °C or by direct use of the element along with addition of TOP can be exploited.



Cadmium acetate (anhydrous, 99.9%) and selenium powder (99%) were purchased from Qualigen India and were used as received. Oleic acid was obtained from Sigma-Aldrich. UV–visible measurements were done in toluene on JASCO V-570 UV-Visible-NIR Spectrophotometer. Photoluminescence spectra were measured in toluene using Hitachi F-2500 Fluorescence Spectrophotometer. SEM was carried out on a Philips XL-30 instrument by dispersing NCs on an aluminium substrate. Powder XRD pattern was obtained on a Mini Flex Rigaku instrument using Cu-Kα radiation (λ = 1.5406 Å). XPS analysis of powdered sample was done on a 9 channeltron CLAM4 analyzer under a vacuum better than 1 × 10−8 Torr, using Mg-Kα radiation with a constant pass energy of 50 eV. Thermo-gravimetric analysis (TGA) was performed on a Mettler Toledo instrument with nitrogen as a purging gas at a scanning rate of 10 °C/min. TEM analysis was carried out on Tecnai-G20 high-resolution transmission electron microscopy (HRTEM) working at 200 kV.

Synthesis of surface-capped CdSe and CdTe

Cadmium acetate (2.46 g) was dissolved in oleic acid (10 mL). After complete dissolution, diphenyl ether was added to make the volume to about 50 mL. To this, was added TOPE (E = Se/Te) prepared from respective element (2 g) in 10 mL of TOP was added under argon. The reaction mixtures were heated to 150 and 200 °C with constant stirring for about 5–8 h to cause an orange to brown precipitation. The orange–red–brown reaction mixture was cooled to room temperature followed by addition of 30–50 mL of toluene and stirring at room temperature for half-an-hour. The reaction mixture was then centrifuged and was washed with the same solvent to obtain orange or brown powder.

Results and discussion

CdSe NCs were prepared by heating cadmium acetate and TOPSe at about 150 and 200 °C whereby continued heating for a period of 5–8 h afforded orange or brown NCs of CdSe by centrifugation method. The reaction of oleic acid with cadmium acetate should essentially lead to formation of an intermediate cadmium oleate by replacing acetate group in the starting material. Cadmium oleate thus generated then reacts with TOPE (E = Se, Te) to form CdE surrounded by oleate group through a sort of Wander Wall’s forces. This method protects the nucleation of the tiny particles. Long carbon chain of the oleic acid coupled with the –C=C– in the middle is considered to be a key factor for better surface capping of the nano-particles (Scheme 1).
Scheme 1

Schematic presentation of formation of CdE

UV–visible absorption spectra of re-dispersed CdSe in toluene showed two absorption bands at the region of 431–34 and 458–60 nm (band gap ~ 2.70 eV) having an edge at about 500 nm (Fig. 1a, b). Observation of absorption band due to first electronic transition indicates that particles have narrow size distribution. The bands in the region of 431–34 nm can originate due to early stage formation of nano-particles and, if isolated, these fall in a different category. If, however, these can be coupled with additional band in the region of 400 nm, these are termed as ‘magic number’ clusters that contain 5–20 atoms. Such CdSe clusters have been recently discussed by Rosenthal et al. (2007) and have been trapped in our laboratory too, and the results of which will be published elsewhere (Khanna et al. 2008 unpublished results).
Fig. 1

UV–visible absorption spectra of CdSe (a) reaction temperature at 150 °C, (b) ~200 °C and (c) UV–visible reflectance spectrum of CdSe prepared at 150 °C

Estimated band width of less than 20 nm further suggested excellent quality of QDs. Generally, a narrow absorption band width is observed for mono-disperse particles with only few electronic defect sites. The band gap of bulk CdSe is reported to be 712 nm (1.74 eV) (Protasenko et al. 2006). Thus, the present measurements show that there is a blue shift of more than 250 nm due to pronounced size quantization effect. Solid-state UV–visible reflectance spectrum showed reflectance wavelength below 500 nm indicating that even in agglomerated state particle retained their nano-dimensions and increased energy-gap values (Fig. 1c). The particle size from the absorption spectra (Hambrock et al. 2001) was calculated by taking advantage of famous effective mass approximation equation (Eq. 1), and the value of 458 or 460 nm resulted in the particle size of about <4 nm.
$$ \Updelta E_{\text{g}} = {{h^{ 2} } \mathord{\left/ {\vphantom {{h^{ 2} } { 8a^{ 2} }}} \right. \kern-\nulldelimiterspace} { 8a^{ 2} }}\left( {{ 1\mathord{\left/ {\vphantom { 1{m_{\text{e}} }}} \right. \kern-\nulldelimiterspace} {m_{\text{e}} }} + { 1\mathord{\left/ {\vphantom { 1{m_{\text{h}} }}} \right. \kern-\nulldelimiterspace} {m_{\text{h}} }}} \right) $$
where Eg is the band-gap shift, a is the particle size and me and mh are the effective electron and hole mass, respectively; me = 0.13m0 and mh = 0.44m0 (data from bulk CdSe), where m0 = 9.1095 × 10−31 kg is the electron rest mass.
The photoluminescence measurement of the same solution showed an emission band at 515 nm (Fig. 2a) at an excitation energy of 400–450 nm. Thus, the emission band was Stoke shifted by more than 50 nm in comparison to absorption wavelength. This slightly large stoke shift can be due to the presence of ‘magic number’ clusters which have distorted geometry and consequently lead to such observations. In addition, the presence of excessive surfactant may disturb the uniformity of the particles. It has been reported that normally an emission energy of about 525–540 nm arises from the particles that have diameter <5 nm (Murray et al. 1993, 2001). Thus from our samples, the emission band at 515 nm can be considered from particles smaller than 5 nm. We believe that a shift in emission wavelength for the sample prepared at 200 °C could be due to relaxation of electron into shallow-trap states, and the origin of emission (Fig. 2b) may be similar to that has been explained by Murray et al. (1993, 2001) for a crystallite of 3–4 nm in size.
Fig. 2

Photoluminescence spectra of CdSe (a) prepared at <150 °C, (b) ~200 °C

Similarly, the reaction of Te-precursor with the pretreated cadmium metal salt with oleic acid led to formation of CdTe nano-particles. Normally, a long reaction time results in the elongated CdTe NCs, but reaction overtime leads to dot-shaped nanocrystals owing to a continuous consumption of the liquid surfactant (Protasenko et al. 2006). The absorption spectra are presented in Fig. 3. The absorption spectra showed well-defined hump-type bands for CdTe particles at 680 and 700 nm for two different preparations. These values were well matched with the reported data on CdTe for the particle size of ~5 nm. Thus, blue shift of about >100 nm relative to the bulk band gap of CdTe (1.50 eV) was observed. The higher absorption values in comparison to bulk CdTe suggest confinement-induced sub-bands (Protasenko et al. 2006).
Fig. 3

UV–visible spectra of CdTe prepared at (a) ~150 °C, (b) ~200 °C

The X-ray diffraction (XRD) pattern (Figs. 4 and 5) showed broad pattern characteristic of cubic crystal structure of bulk CdSe and CdTe. All peaks of Bragg’s scattering at 111, 220, 311, 400, 331 crystal planes were observed in the diffractogram, and the broad patterns were supportive of small particle size. It was interesting to note that reaction carried out at about 200 °C, yielded cubic (zinc blende) CdSe but hexagonal (wurtzite) CdTe. This observation was similar to that has been described earlier (Murray et al. 1993, 2001). Calculation of crystallite size from line broadening using Scherrer’s equation gave an average crystallite size of <10 nm. Figures 6, 7, 8 and 9 show TEM and SEM micrographs of capped CdSe and CdTe NCs. It is observed that the CdSe particles were highly agglomerated but CdTe nano-particles appeared to be quite well defined in their shapes and size.
Fig. 4

XRD of CdSe (a) reaction temperature <150 °C, (b) ~200 °C
Fig. 5

XRD of CdTe (a) reaction temperature ~150 °C, (b) ~200 °C
Fig. 6

TEM of CdSe
Fig. 7

TEM (a, b) and selected area electron diffraction pattern (c) of CdTe prepared at ~150 °C having cubic phase structure
Fig. 8

TEM (a, b) and electron diffraction pattern recorded with an electron-beam wavelength of 2.5 pm and a camera length of 300 mm (c) of CdTe prepared at 200 °C
Fig. 9

SEM of CdTe prepared at 200 °C

TEM of CdSe showed that the individual particles were stacked over several other similar particles and they appeared as a cluster. However, this was not the case for CdTe. The spherical nano-particles of CdTe were well dispersed and were not so much agglomerated. In fact the high-resolution TEM of CdTe prepared at 150 or 200 °C showed good crystallinity, and the electron diffraction showed that the concentric rings are different for the sample that was prepared at 150 °C than the one prepared at 200 °C. The XRD has shown that the two samples have different crystal structures, so this was also confirmed by electron diffraction pattern and HRTEM. The average size of the particles in sample of CdTe prepared at 150 °C was 5 nm. HRTEM bright field image showed nearly spherical NCs. The HRTEM image of the particles showed lattice fringes with the inter-fringe distance measured to be 0.379 nm, close to the lattice spacing of (111) planes at 0.374 nm in the cubic-structured CdTe. HRTEM image clearly indicated that no crystal defects such as point defects, line defects and stacking faults were present in CdTe QDs synthesized at 150 °C. The selected area electron diffraction (SAED) pattern showed successive inter-planar distances corresponding to the zinc blend (cubic) structure. Similarly, the average particle size in the case of samples synthesized at 200 °C was 9 nm. The lattice planes were visualized within the crystal structure. The inter-planar spacing in CdTe was measured to be 0.406 nm, which corresponds to the characteristic (100) lattice plane in hexagonal (wurtzite) CdTe (0.398 nm). The SAED pattern showed successive inter-planar distances for wurtzite crystal structure. Figure 8 shows a SAED pattern recorded at an accelerating voltage of 200 kV. The diffuse ring immediate to the transmitted beam in the SAED pattern in Fig. 8 is indexed as the (102) plane. The second, third and fourth rings in Fig. 8 correspond to (103), (203) and (110) planes, respectively. If CdTe in the present synthesis procedure crystallized in zinc blende structure, then no ring corresponding (102) plane would be observed in the SAED pattern. The appearance of (102) plane confirms that CdTe crystallizes in wurtzite structure at 200 °C.

The composition of CdTe was qualitatively determined by means of energy disperse X-ray analysis (EDS) measurements. The ratio of Cd:Te was found to be 1:1.06, confirming correct ratio of the two elements. SEM of CdTe showed that the particles were well scattered but were aggregated due to presence of surfactant.

A representative IR spectrum of CdSe recorded in KBr matrix shows typical peaks for presence of organics (Fig. 10). Peaks between ν1000–1800 cm−1 and ν2800–3000 cm−1 are attributable to the symmetric and asymmetric stretching of the –CH2 groups, terminal –CH3 and =CH of oleic acid. A peak in the close vicinity of ν1450 is considered due to CH2 deformation vibration. Similarly, for oleic acid, a band in the vicinity of ν1700–1720 cm−1 is anticipated due to the C=O vibration (Wu et al. 2004).
Fig. 10

IR spectrum of oleic acid-capped CdSe

TGA (Fig. 11) showed a three-stage decomposition pattern for the CdSe NCs capped with oleic acid. The presence of TOP cannot be ruled out in the final sample. Therefore, first two decompositions at about 250–330 °C can be considered due to vapourization of organic capping (oleic acid and TOP) from the surface of the NCs. The third-stage weight loss at about 800 °C is considered due to decomposition of CdSe. A total of about 35% wt loss were observed when the analysis is done between room temperature and 900 °C at a rate of 10 °C/min under nitrogen atmosphere.
Fig. 11

TGA of oleic acid-capped CdSe

In order to examine the chemical structure of surface-coated CdSe, XPS spectra (Fig. 12) of Cd(3d) and Se(3d) were obtained. The spectra showed two peaks at orbit splitting of Cd 3d5/2–Cd 3d3/2 at about 405 and 411.7 eV and one peak at 54.0 eV for Se 3d binding energy in CdSe (Fig. 12a). The presence of single peak for Se confirmed that the CdSe NCs were effectively capped by oleic acid because additional peak due to SeO2 on the surface was not observed. However, in case of CdTe an additional doublet for Te 3d levels was clearly seen. The doublet in Te spectrum indicated that the surface of the nano-particles of CdTe was oxidation prone. Therefore, binding energies of Te 3d5/2 peaks were observed at 572.2 and 576.1 eV (Fig. 12b). Based on previously published XPS results on CdTe, the peak at 572.2 eV was assigned to Te–Cd bonds in CdTe and peak at 576.1 eV Te–O bonds in TeO2 which might form due to surface oxidation of the nano-particles because of the poor capping or even due to prolonged exposure to normal environment (Briggs and Seah 1996; Pei et al. 1998). The binding energy of 405.1 eV for Cd 3d5/2 signal agrees very well with the reported values for Cd bonded to Te in CdTe (Briggs and Seah 1996; Pei et al. 1998). We conducted the XPS studied at the very end after all data collection was over and this took some time; therefore, we opine that sample might have got slightly oxidized due to environmental exposure.
Fig. 12

XPS binding energy analysis of CdSe (a) and CdTe (b) capped with oleic acid


Re-dispersible CdSe and CdTe nano-particles have been synthesized with effective capping of oleic acid in presence of TOP. Optical studies showed that the absorption and emission properties are well matched with the reported values for the particle diameter of less than 5 nm. A blue shift of more than 250 nm was found in CdSe indicating an increased band gap of so-prepared material. Formation of cubic crystal structure was predominant at 150 °C for both; however, at 200 °C hexagonal CdTe was obtained as against cubic CdSe. XPS analysis showed no peak due to SeO2 indicating that the particles are stable for a long time. The current work highlights the usefulness of one-pot synthesis of semiconductor QDs and allows the researchers to adopt the method for their day-to-day research in nano-technology.


PKK thanks Dr. T. L. Prakash, Director, C-MET, Hyderabad, for permitting KRSR to work in his laboratory at Pune and DST (Government of India) for financial support through grant no. SR/S1/PC-17/2006.

Copyright information

© Springer Science+Business Media B.V. 2009