In situ growth of well-dispersed CdS nanocrystals in semiconducting polymers
© Laera et al.; licensee Springer. 2013
Received: 19 May 2013
Accepted: 2 August 2013
Published: 9 September 2013
Skip to main content
© Laera et al.; licensee Springer. 2013
Received: 19 May 2013
Accepted: 2 August 2013
Published: 9 September 2013
A straight synthetic route to fabricate hybrid nanocomposite films of well-dispersed CdS nanocrystals (NCs) in poly[2-methoxy-5-(2'-ethyl-hexyloxy)-1,4-phenylene vinylene] (MEH-PPV) is reported. A soluble cadmium complex [Cd(SBz)2]2·MI, obtained by incorporating a Lewis base (1-methylimidazole, MI) on the cadmium bis(benzyl)thiol, is used as starting reagent in an in situ thermolytic process. CdS NCs with spherical shape nucleate and grow well below 200°C in a relatively short time (30 min). Photoluminescence spectroscopy measurements performed on CdS/MEH-PPV nanocomposites show that CdS photoluminescence peaks are totally quenched inside MEH-PPV, if compared to CdS/PMMA nanocomposites, as expected due to overlapping of the polymer absorption and CdS emission spectra. The CdS NCs are well-dispersed in size and homogeneously distributed within MEH-PPV matrix as proved by transmission electron microscopy. Nanocomposites with different precursor/polymer weight ratios were prepared in the range from 1:4 to 4:1. Highly dense materials, without NCs clustering, were obtained for a weight/weight ratio of 2:3 between precursor and polymer, making these nanocomposites particularly suitable for optoelectronic and solar energy conversion applications.
Efficient photoconductivity requires not only efficient charge separation but also efficient transport of the carriers to the electrodes without recombination, in that sense, the morphology of nanocomposite being crucial in providing suitable paths for both electron and hole towards the appropriate electrode . The NC network must be homogeneous so that each negative charge can efficiently hop to another NC in the direction of the internal field, this requirement being a complex issue when NCs are dispersed in polymeric matrices. The main difficulty is due to the high surface-to-volume ratio of NCs that tend to form agglomerate to lower their surface energy. Furthermore, the addition of a dense network of NCs to polymers can significantly alter the mechanical properties of the resulting nanocomposite material compromising the advantageous properties of organic semiconductor such as the easy processability .
The nanocomposite is frequently gained by solution blending, i.e. dispersion of NCs in polymer solutions that can be dried under vacuum or can be used to obtain thin films by spin-casting (solvent evaporation) . During these procedures, the NCs form microsized aggregates and cannot be separated from each other. As a consequence, nanocomposites have been commonly prepared by synthesis of the inorganic NCs in situ, for instance in solution, where the solvent is a monomer and the nanocomposite is then prepared through in situ polymerization [11, 12]. Alternatively, the inorganic NCs can be synthesized inside polymer matrices through the thermolysis of suitable precursors. Recent works of our research group have demonstrated that cadmium thiolates are promising materials for the in situ synthesis of nanocrystalline CdS –. Using unimolecular precursors, as cadmium thiolates, it is possible to overcome any problem, occurring in the other chemical methods, such as the low temporal stability of reagents, the inhomogeneity of multicomponent mixing and the intrinsic high reactivity and toxicity of the precursor used. Furthermore, unimolecular precursors guarantee the stoichiometry control of thermolytic process. Unfortunately, cadmium thiolates, having a polymeric structure, are insoluble in typical organic solvents; so, it is not possible to homogeneously disperse them in polymeric matrices, and the thermolysis process induces the growth of CdS NCs with a disordered distribution.
In this work, we present a straight convenient approach to synthesize well-dispersed CdS NCs in pristine MEH-PPV using a cadmium complex obtained by incorporating a Lewis base (1-methylimidazole, MI) on the cadmium bis(benzyl)thiol , the [Cd(SBz)2]2·MI adduct (for the sake of brevity CBzMI). The latter proves to be highly soluble in the most common organic solvents. Solutions of the polymer MEH-PPV and the cadmium complex allow to obtain large area composite films by spin coating, making the proposed technique not expensive and ideal to fabricate optoelectronic devices.
All the reagents used to synthesize the precursor and the polymer were purchased from Sigma-Aldrich S.r.l., Milan, Italy, and used without further purification. All the nanocomposites were prepared using the pristine polymer MEH-PPV with a number of average molecular weight (Mn) of 70,000 to 100,000.
The synthesis of Cd(SBz)2 was conducted using the commercial salt cadmium nitrate hexahydrate (9 mmol) as starting reagent. After the dissolution of cadmium salt in ethanol, an aqueous solution of ammonium hydroxide (25%) was added and, as a consequence, the starting opaque solution became clear. When the benzyl mercaptan (18 mmol) was added in the reaction vessel, the desired product precipitated in quantitative yield and it was isolated from the solution by filtration.
The soluble complex [Cd(SBz)2]2·MI was performed suspending the thiolate Cd(SBz)2 and adding dropwise 1-methyl imidazole (MI) until a clear solution was obtained. The product was purified by crystallization from toluene, cooling the solution to −18°C. Thermogravimetric analysis (Netzsch-Gerätebau GmbH STA429 simultaneous thermal analyzer, Selb, Germany) allowed to confirm the general formula of the obtained Lewis base-derived complex [Cd(SBz)2]2·MI in which the stoichiometric ratio between thiolate and MI is 2:1 .
The precursor/polymer composite films were produced by spin coating on glass slides, silicon wafers and copper grids from the solutions of [Cd(SBz)2]2.MI and MEH-PPV in chloroform with a respective weight/weight ratio of 1:4, 2:3 and 4:1, respectively. The same procedure was realized using an inert polymer as polymethyl methacrylate (PMMA) for comparative aims. The spin speed and time were set at 1,500 rpm and 10 s, respectively, in order to obtain uniform and smooth polymer films. For all samples, the thermolysis process was performed at temperatures of 175°C, 185°C and 200°C for 30 min with a reproducible controlled ramp and in nitrogen atmosphere to avoid possible oxidation of NCs surface.
Optical properties of the annealed samples, by means of a Xe lamp (LC8 Hamamatsu, Hamamatsu City, Shizuoka, Japan) and a HR460 monochromator (Jobin Yvon, Kyoto, Japan), were investigated on chloroform solutions obtained by the samples deposited on glass. UV-visible transmission were performed in order to evaluate the absorbance of the specimens as ln(1/T). Photoluminescence (PL) spectra were acquired on the same chloroform solutions with a Varian Cary Eclipse Fluorometer, Palo Alto, CA, USA, (excitation wavelength, 330 nm).
In order to investigate the formation of CdS NCs, their structural properties, crystallographic structure and size, X-ray diffraction measurements were performed on the annealed samples on glass substrates using a Philips PW1880 X-ray diffractometer, Almelo, The Netherlands, having a CuKα radiation source (3 kW).
Using a TECNAI F30 transmission electron microscope (TEM), FEI, Hillsboro, OR, USA, operating at 300 kV and point-to-point resolution of 0.205 nm, the structural characterization of the samples deposited on carbon-coated copper grids was also executed. Finally, rheological measurements were carried out by a parallel plate rheometer stress tech HR at 200°C. Samples of MEH-PPV and CdS/MEH-PPV nanocomposites, with a relative weight ratio of 1:4, were prepared by casting of solution in chloroform to obtain 1-mm thick films in order to evaluate the influence of CdS NCs inclusion on MEH-PPV film mechanical properties.
CdS NC size calculated from absorption data
Annealing temperature (°C)
Absorption edge (nm)
Band gap absorption (eV)
CdS NC size (from Brus equation []) (nm)
Curve B in Figure 5 shows the WAXS pattern of the CdS/MEH-PPV nanocomposites obtained after annealing at 185°C for the samples with a weight/weight ratio of 1:4. Here, besides the MEH-PPV diffraction peaks, broad X-ray peaks attributed to the formation of CdS nanocrystals are also observed. Also, curve C obtained for the samples with a weight/weight ratio of 4:1 shows the CdS nanocrystal peaks. However, in this case, the polymer peaks (P and the weak peaks of the polymer superstructure) are not observed or are too low to be experimentally observed due to the low polymer content.
In order to analyse the structure and particle size of the CdS nanocrystals, the experimental WAXS pattern of the sample with a weight/weight ratio of 1:4 (curve C) is compared to the simulated X-ray diffraction patterns for cubic (zinc blende) and hexagonal (wurtzite) CdS nanocrystals. For the sake of simplicity, here, we focus our comparison to curve C because in curve B, the polymer peak P is overlapped to the main CdS diffraction peak, but as can be easily seen, the conclusion and findings will be identical for curve B.
For this kind of polymer nanocomposite samples, it is not very easy to perform quantitative X-ray analyses; nevertheless, by comparing the calculated patterns with experimentally measured patterns, we find a much better agreement for the wurtzite phase of the CdS nanocrystals. This is particularly evident for the shape of the main diffraction peak (convolution of more Bragg peaks) at about 2θ = 27.6° and for the broad peak at about 2θ = 47°. Nevertheless, we cannot exclude the presence and coexistence of CdS nanocrystals of zinc blende phase within the hybrid nanocomposite.
These results indicate that the inclusion of NCs into the polymer matrix does not significantly alter the polymer resistance to deformation. Applications in the field of large-area, flexible, low-cost solar cells require to preserve the nanoflow material rheology to allow the developing of fabrication process based on spinning or soft moulding lithography.
The rheological measurements complete the characterization of prepared CdS/MEH-PPV hybrid nanocomposites. All data acquired by absorption spectroscopy, X-ray diffraction and TEM show the growth of CdS NCs with a regular spherical shape, a narrow size distribution and a homogenous dispersion inside the polymer. The use of 1-methylimidazole ligand to improve the solubility of Cd(SBz)2 has allowed to obtain clear solutions of the complex [Cd(SBz)2]2·MI and MEH-PPV in chloroform, suitable to prepare thin solid film using the cheap and easy technique of spin coating. The CdS NCs size grows from 2.8 to 3.5 nm in the temperature range 175°C to 200°C, demonstrating a slow and controlled diffusion of Cd(SBz)2 molecules inside the matrix. Nevertheless, the polymer prevents the formation of microsize aggregate by limiting the diffusion capability of CdS NCs.
The presented synthetic strategy allows a good control of NC size and distribution within the polymer matrix as required for the application in photovoltaic cells.
An in situ synthetic route for the realization of hybrid polymer/nanocomposite materials was presented. We demonstrated that the soluble metal thiolate derivative [Cd(SBz)2]2·MI, obtained using 1-methylimidazole as cadmium ligand, is a suitable starting material to grow CdS NCs in semiconducting polymeric matrices. We found that the precursor decomposition and the subsequent NCs nucleation and growth start at temperatures below 200°C, namely already at 175°C and in relatively short time (30min), the temperature lowering being crucial for avoiding possible damage or deterioration of the matrix. Such a result allows extending the range of suitable matrices to thermally soft polymers such as MEH-PPV towards the fabrication of organic–inorganic nanocomposite materials for optoelectronics and light harvesting.
The structure of [Cd(SBz)2]2·MI also helps in obtaining a homogeneous spatial dispersion of the molecule itself inside the polymer promoting the formation of a highly uniform network and well-dispersed NCs. The weight ratio of the precursor to the polymer directly determines the number density of the NCs as well as the coverage uniformity, the optimal value being 2:3.
The synthetic route did not significantly alter the polymer resistance to deformation, further demonstrating the applicability in the field of large-area, flexible, low-cost solar cells production via spinning or soft moulding lithography.
This work was supported by the Regione Puglia (Bari, Italy) - Project PONAMAT (PS_016).
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.