Wet Chemical Synthesis of Highly Luminescent HgTe/CdS Core/Shell Nanocrystals

Abstract

We have recently reported the wet chemical synthesis of colloidal HgTe nanocrystals in aqueous solution at room temperature. This novel material exhibits very strong and broad photoluminescence (PL) in the near infrared, with the exact wavelength depending on the synthetic conditions. The preparation was an extension to the previously published work on cadmium chalcogenide nanocrystals, i.e., CdS, CdSe, and CdTe, and used 1-thioglycerol as the size-regulating capping agent. Particles with a large distribution of sizes in the 3±6 nm diameter range were produced. Luminescence quantum efficiencies (QEs) of around 50 %, which are among the highest ever reported for a colloidal system, and stability towards oxidation indicate that the surface of the HgTe nanocrystals is extremely well-passivated. Typically, the PL from the freshly prepared material is around 1050 nm, but this moves to longer wavelengths with time, accompanied by an apparent drop in QE, until after two weeks the luminescence has shifted out to around 1200 nm. This aagingo process continues indefinitely, albeit at a slower rate, and presents real cause for concern for the long-term stability of such materials in device applications. In addition, if the colloidal nanocrystal solutions are heated, then a similar but more rapid process occurs with the PL shifting to beyond 1500 nm and the QE dropping to < 1 % with just a few minutes refluxing at 100 C. Again, this lack of stability to heating could be problematic in certain applications if high optical pump powers are used in laser or amplifier devices. A similar, although less-pronounced, effect is also observed in 1-thioglycerol-stabilized CdTe nanoparticles, which undergo a red shift of the PL under prolonged reflux. Whether this shift is caused by a gradual aripeningo of the nanocrystals with heat and/or time, or actually represents a compositional change will be discussed at a later date. Needless to say, the effect is undesirable and for HgTe to be a useful material, operating at the strategic near-infrared telecommunications wavelengths, the problem has to be eliminated. A possible way of achieving this is to cap the surface of the nanocrystals with a higher bandgap inorganic layer. These acore/shello composite materials can increase the luminescence quantum yield due to improved passivation of the surface, and also tend to be more physically robust than the abareo organically passivated clusters. This should therefore produce a much more stable material where not only the aagingo process is suppressed, but which is more tolerant to the processing conditions necessary for incorporation into useful devices. Examples of such core/shell structures include CdSe/ZnS, CdS/ZnS, CdSe/CdS, CdSe/ZnSe, and CdS/HgS with reports of roomtemperature QEs of up to 50 % for the CdSe/ZnS material. It was therefore decided to attempt to overcoat our HgTe nanocrystals with a layer of CdS in order to produce a physically stable core/shell heterostructure. The abareo 1-thioglycerol-stabilized HgTe colloidal nanocrystals were prepared in aqueous solution at room temperature using the method described previously. In order to coat these abareo organically capped nanocrystals with a layer of CdS, H2S gas buffered in N2 was passed through a vigorously stirred, dilute aqueous solution of HgTe nanocrystals and cadmium(II) perchlorate at a pH of around 10 in the presence of extra 1-thioglycerol stabilizer. The mixture was initially slightly turbid, but this cleared upon injection of the H2S, resulting in a clear, goldenbrown solution. This solution was then refluxed for about 30 min as a check of the robustness of the HgTe/CdS system. In order to fully characterize the material, optical absorption and photoluminescence spectra were recorded at all stages of the synthesis, as well as X-ray diffraction (XRD) patterns and high-resolution transmission electron microscopy (HRTEM) images to confirm the presence of HgTe/CdS core/shell particles in the final solution. The optical absorption data is shown in Figure 1, comparing the freshly prepared abareo HgTe prior to capping, and the subsequent HgTe/CdS material. A clear aexcitonico peak is visible for the bare HgTe sample at 850 nm, which clearly shifts to the red when the CdS capping is introduced. The increased optical density at shorter wavelengths could be indicative of some discrete CdS or HgCdTeS alloyed nanoparticles being formed in addition to the core/shell material. The effect of aging and/or heating the samples, both capped and uncapped, is to red-shift and broaden the absorption edges into the region where water absorption caused by a slight cell-mismatch prevents the recording of good-quality spectra. These are, therefore,