Limits...
Rational design of the gram-scale synthesis of nearly monodisperse semiconductor nanocrystals.

Protière M, Nerambourg N, Renard O, Reiss P - Nanoscale Res Lett (2011)

Bottom Line: On the basis of these results, the synthesis has been scaled up by a factor of 20.Using a 2-L batch reactor combined with a high-throughput peristaltic pump, different-sized samples of CdSe nanocrystals with yields of 2-3 g per synthesis have been produced without sacrificing the narrow size distribution.In a similar setup, the gram-scale synthesis of CdSe/CdS/ZnS core/shell/shell nanocrystals exhibiting a fluorescence quantum yield of 81% and excellent resistance of the photoluminescence in presence of a fluorescent quencher (aromatic thiol) has been achieved.PACS: 81.20.Ka, 81.07.Bc, 78.67.Bf.

View Article: PubMed Central - HTML - PubMed

Affiliation: DSM/INAC/SPrAM (UMR 5819 CEA-CNRS-UJF)/LEMOH, CEA-Grenoble - 17 rue des Martyrs - 38054 Grenoble cedex 9, France. peter.reiss@cea.fr.

ABSTRACT
We address two aspects of general interest for the chemical synthesis of colloidal semiconductor nanocrystals: (1) the rational design of the synthesis protocol aiming at the optimization of the reaction parameters in a minimum number of experiments; (2) the transfer of the procedure to the gram scale, while maintaining a low size distribution and maximizing the reaction yield. Concerning the first point, the design-of-experiment (DOE) method has been applied to the synthesis of colloidal CdSe nanocrystals. We demonstrate that 16 experiments, analyzed by means of a Taguchi L16 table, are sufficient to optimize the reaction parameters for controlling the mean size of the nanocrystals in a large range while keeping the size distribution narrow (5-10%). The DOE method strongly reduces the number of experiments necessary for the optimization as compared to trial-and-error approaches. Furthermore, the Taguchi table analysis reveals the degree of influence of each reaction parameter investigated (e.g., the nature and concentration of reagents, the solvent, the reaction temperature) and indicates the interactions between them. On the basis of these results, the synthesis has been scaled up by a factor of 20. Using a 2-L batch reactor combined with a high-throughput peristaltic pump, different-sized samples of CdSe nanocrystals with yields of 2-3 g per synthesis have been produced without sacrificing the narrow size distribution. In a similar setup, the gram-scale synthesis of CdSe/CdS/ZnS core/shell/shell nanocrystals exhibiting a fluorescence quantum yield of 81% and excellent resistance of the photoluminescence in presence of a fluorescent quencher (aromatic thiol) has been achieved.PACS: 81.20.Ka, 81.07.Bc, 78.67.Bf.

No MeSH data available.


Related in: MedlinePlus

Absorption, PL spectra, and TEM images of the three samples obtained. UV-vis absorption (a) and normalized photoluminescence (b) spectra of the three different-sized samples (excitation wavelength, 400 nm). (c) TEM images: 3.6-, 4.5-, and 6.1-nm CdSe nanocrystals (from left to right) at identical magnification and high resolution images of single particles (image size 7 × 7 nm).
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC3211985&req=5

Figure 1: Absorption, PL spectra, and TEM images of the three samples obtained. UV-vis absorption (a) and normalized photoluminescence (b) spectra of the three different-sized samples (excitation wavelength, 400 nm). (c) TEM images: 3.6-, 4.5-, and 6.1-nm CdSe nanocrystals (from left to right) at identical magnification and high resolution images of single particles (image size 7 × 7 nm).

Mentions: The results of the DOE analysis point at the combination A2, B1, C1, D1, E1, F2, and G2 for obtaining the smallest size of nanocrystals. However, stearic acid was preferred over oleic acid with the goal to minimize the synthesis cost (level F = 1). The experiment with the combination A2, B1, C1, D1, E1, F1, and G2 led indeed to small nanocrystals (3.3 nm diameter) but the size dispersion was unacceptable as indicated by the FWHM (43 nm). With the goal to decrease the size dispersion, the parameters D and E were set to level 2, as (1) the interaction DF has a strong influence on the FWHM and (2) increasing the ligand/Cd ratio tends to lower the FWHM (vide supra). The resulting combination (A2, B1, C1, D2, E2, F1, and G2) has already been tested (trial 10) and led to rather large nanocrystals (5.2 nm) having an important size dispersion (FWHM 46 nm). These results indicate that the experimental domain for one or several parameters have been chosen too narrow in view of obtaining low size distributions. The molar quantity of oleylamine is a parameter, which had been - in contrast to all other parameters - rather arbitrarily set to 25% of the molar quantity of the stabilizing ligand (stearic or oleic acid), due to the absence of appropriate literature results. Therefore the value fixed in the experimental plan may lie outside the optimum range. Using the same molar quantity of oleylamine and stearic acid (10 mmol) led to nanocrystals with a size of 3.6 nm ("small size") and a FWHM of 28-29 nm. Alkylamines are known to "activate" metal carboxylate complexes in the synthesis of II-VI and III-V nanocrystals [30,31]. Therefore, in order to obtain larger nanocrystals ("medium size"), the quantity of oleylamine has been significantly increased to 42.5 mmol, resulting in particles with a diameter of 4.5 nm and an average FWHM of 27.5 nm. The reaction yields were 95% (3.6 nm) and 80% (4.5 nm), i.e., in both cases the conversion of the minority precursor (Cd) is close to quantitative. The synthesis of even bigger nanocrystals ("large size") is based on the parameters used in experiment 8, which resulted in 5.5-nm particles with a FWHM of 30 nm. Increasing at the same time the cadmium concentration and the ligand:Cd ratio by a factor of 2 led to 6.1-nm nanocrystals (FWHM 30 nm). However, working with a large ligand/Cd ratio of 50:1 in the absence of oleylamine reduced the reactivity and conversion of the cadmium precursor and therefore the reaction yield was only 60% in this case. The absorption and PL spectra as well as TEM images of the three samples obtained are depicted in Figure 1.


Rational design of the gram-scale synthesis of nearly monodisperse semiconductor nanocrystals.

Protière M, Nerambourg N, Renard O, Reiss P - Nanoscale Res Lett (2011)

Absorption, PL spectra, and TEM images of the three samples obtained. UV-vis absorption (a) and normalized photoluminescence (b) spectra of the three different-sized samples (excitation wavelength, 400 nm). (c) TEM images: 3.6-, 4.5-, and 6.1-nm CdSe nanocrystals (from left to right) at identical magnification and high resolution images of single particles (image size 7 × 7 nm).
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC3211985&req=5

Figure 1: Absorption, PL spectra, and TEM images of the three samples obtained. UV-vis absorption (a) and normalized photoluminescence (b) spectra of the three different-sized samples (excitation wavelength, 400 nm). (c) TEM images: 3.6-, 4.5-, and 6.1-nm CdSe nanocrystals (from left to right) at identical magnification and high resolution images of single particles (image size 7 × 7 nm).
Mentions: The results of the DOE analysis point at the combination A2, B1, C1, D1, E1, F2, and G2 for obtaining the smallest size of nanocrystals. However, stearic acid was preferred over oleic acid with the goal to minimize the synthesis cost (level F = 1). The experiment with the combination A2, B1, C1, D1, E1, F1, and G2 led indeed to small nanocrystals (3.3 nm diameter) but the size dispersion was unacceptable as indicated by the FWHM (43 nm). With the goal to decrease the size dispersion, the parameters D and E were set to level 2, as (1) the interaction DF has a strong influence on the FWHM and (2) increasing the ligand/Cd ratio tends to lower the FWHM (vide supra). The resulting combination (A2, B1, C1, D2, E2, F1, and G2) has already been tested (trial 10) and led to rather large nanocrystals (5.2 nm) having an important size dispersion (FWHM 46 nm). These results indicate that the experimental domain for one or several parameters have been chosen too narrow in view of obtaining low size distributions. The molar quantity of oleylamine is a parameter, which had been - in contrast to all other parameters - rather arbitrarily set to 25% of the molar quantity of the stabilizing ligand (stearic or oleic acid), due to the absence of appropriate literature results. Therefore the value fixed in the experimental plan may lie outside the optimum range. Using the same molar quantity of oleylamine and stearic acid (10 mmol) led to nanocrystals with a size of 3.6 nm ("small size") and a FWHM of 28-29 nm. Alkylamines are known to "activate" metal carboxylate complexes in the synthesis of II-VI and III-V nanocrystals [30,31]. Therefore, in order to obtain larger nanocrystals ("medium size"), the quantity of oleylamine has been significantly increased to 42.5 mmol, resulting in particles with a diameter of 4.5 nm and an average FWHM of 27.5 nm. The reaction yields were 95% (3.6 nm) and 80% (4.5 nm), i.e., in both cases the conversion of the minority precursor (Cd) is close to quantitative. The synthesis of even bigger nanocrystals ("large size") is based on the parameters used in experiment 8, which resulted in 5.5-nm particles with a FWHM of 30 nm. Increasing at the same time the cadmium concentration and the ligand:Cd ratio by a factor of 2 led to 6.1-nm nanocrystals (FWHM 30 nm). However, working with a large ligand/Cd ratio of 50:1 in the absence of oleylamine reduced the reactivity and conversion of the cadmium precursor and therefore the reaction yield was only 60% in this case. The absorption and PL spectra as well as TEM images of the three samples obtained are depicted in Figure 1.

Bottom Line: On the basis of these results, the synthesis has been scaled up by a factor of 20.Using a 2-L batch reactor combined with a high-throughput peristaltic pump, different-sized samples of CdSe nanocrystals with yields of 2-3 g per synthesis have been produced without sacrificing the narrow size distribution.In a similar setup, the gram-scale synthesis of CdSe/CdS/ZnS core/shell/shell nanocrystals exhibiting a fluorescence quantum yield of 81% and excellent resistance of the photoluminescence in presence of a fluorescent quencher (aromatic thiol) has been achieved.PACS: 81.20.Ka, 81.07.Bc, 78.67.Bf.

View Article: PubMed Central - HTML - PubMed

Affiliation: DSM/INAC/SPrAM (UMR 5819 CEA-CNRS-UJF)/LEMOH, CEA-Grenoble - 17 rue des Martyrs - 38054 Grenoble cedex 9, France. peter.reiss@cea.fr.

ABSTRACT
We address two aspects of general interest for the chemical synthesis of colloidal semiconductor nanocrystals: (1) the rational design of the synthesis protocol aiming at the optimization of the reaction parameters in a minimum number of experiments; (2) the transfer of the procedure to the gram scale, while maintaining a low size distribution and maximizing the reaction yield. Concerning the first point, the design-of-experiment (DOE) method has been applied to the synthesis of colloidal CdSe nanocrystals. We demonstrate that 16 experiments, analyzed by means of a Taguchi L16 table, are sufficient to optimize the reaction parameters for controlling the mean size of the nanocrystals in a large range while keeping the size distribution narrow (5-10%). The DOE method strongly reduces the number of experiments necessary for the optimization as compared to trial-and-error approaches. Furthermore, the Taguchi table analysis reveals the degree of influence of each reaction parameter investigated (e.g., the nature and concentration of reagents, the solvent, the reaction temperature) and indicates the interactions between them. On the basis of these results, the synthesis has been scaled up by a factor of 20. Using a 2-L batch reactor combined with a high-throughput peristaltic pump, different-sized samples of CdSe nanocrystals with yields of 2-3 g per synthesis have been produced without sacrificing the narrow size distribution. In a similar setup, the gram-scale synthesis of CdSe/CdS/ZnS core/shell/shell nanocrystals exhibiting a fluorescence quantum yield of 81% and excellent resistance of the photoluminescence in presence of a fluorescent quencher (aromatic thiol) has been achieved.PACS: 81.20.Ka, 81.07.Bc, 78.67.Bf.

No MeSH data available.


Related in: MedlinePlus