Synthesis of high-quality semiconductor nanoparticles in a composite hot-matrix

of 4

Please download to get full document.

View again

All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
PDF
4 pages
0 downs
0 views
Share
Description
Semiconductor nanocrystals are of a great interest for many practical applications which motivates the search of low cost and environmental-friendly methods for their manufacturing. Here we report the synthesis of CdSe and CdS nanoparticles utilizing
Tags
Transcript
  Introduction Semiconductor nanoparticles (quantum dots, QDs) areapplied in photovoltaics, LEDs and lasers, diagnosticlabels in medicine, etc. They require an increasingamount of high-quality nanoparticles synthesized bycost-effective and environmental-friendly methods. Inthe previous hot-matrix methods [1, 2] an organome- tallic precursor, e.g., dimethyl cadmium, is employed forthe preparation of semiconductor nanoparticles. How-ever, the organometallic precursors are expensive, diffi-cult for synthesis and very toxic. The usage of CdO asCd precursor [3] in a molten coordinating matrix of trioctylphosphine oxide (TOPO) is the most preferredway, because of nonuse of organometallics. However,the use of such coordinating matrixes is restrictive forthe large-scale production of nanoparticles, because of expensive regeneration cost of the matrix and environ-mental pollution at disposal. Although there are suitableliquids and surfactants for the low-temperature micro-emulsion synthesis [4], the monodispersity of nanopar-ticles is rather low. In this study we report a modifiedprocedure for the synthesis of A II B VI type semicon-ductor nanoparticles by a non-coordinating matrix of liquid paraffin and stearic acid as a coordinating ligand.A procedure, using octadecene and oleic acid is knownfrom the literature [5]. In our case we have chosen thecouple liquid paraffin (matrix solvent) and stearic acid(ligand), because they are low-cost and environmental-friendly materials. Moreover, we obtain high-qualityCdSe and CdS nanoparticles using this method. Fur-thermore, we compared the two kinetics of their growthat 250   C in liquid paraffin and in TOPO in order toevaluate both matrices. Georgi G. YordanovCeco D. DushkinGospodinka D. GichevaBocho H. BochevEiki Adachi  Synthesis of high-quality semiconductor nanoparticles in a composite hot-matrix  Received: 23 December 2004Accepted: 9 June 2005Published online: 4 August 2005   Springer-Verlag 2005 Abstract  Semiconductor nanocrys-tals are of a great interest for manypractical applications which moti-vates the search of low cost andenvironmental-friendly methods fortheir manufacturing. Here we reportthe synthesis of CdSe and CdSnanoparticles utilizing compositematrix of liquid paraffin as a non-coordinating solvent and stearic acidas a coordinating ligand. The nano-particle growth kinetics is comparedto that of the classical synthesis intrioctylphosphine oxide matrix. It isfound that the nucleation and crystalgrowth are remarkably affected bythe coordinating ligand. The CdSeand CdS nanocrystals can be iso-lated and purified from the matrixwhich makes it possible their large-scale synthesis for applications. Keywords  CdSe nanocrystals  Æ  CdSnanocrystals  Æ  Liquid paraffinmatrix  Æ  Stearic acid  Æ  Nanoparticlegrowth kinetics Colloid Polym Sci (2005) 284: 229–232DOI 10.1007/s00396-005-1370-x  SHORT COMMUNICATION G. G. Yordanov ( & )  Æ  C. D. DushkinG. D. Gicheva  Æ  B. H. BochevFaculty of Chemistry,Laboratory of Nanoparticle Scienceand Technology,Department of General and InorganicChemistry, University of Sofia,1 James Boucher Blvd., Sofia, 1164,BulgariaE. AdachiAdvanced Research Center,Nihon L’Oreal R & D, 3-2-1 Sakado,Takatsu, Kawasaki, Kanagawa 213-0012,Japan  Experimental procedure and results Three materials, 50 mg (0.38 mmol) of CdO, 15 ml of liquid paraffin and 2 g (7.0 mmol) of stearic acid, are putin a Schlenk flask (50 ml in volume). Then argon gas isblown through the flask mounted on a silicon oil bath ona magnetic stirrer and a flask heater (argon is notabsolutely necessary prerequisite). Though the color of the solution mixture in the flask is initially brown, itbecomes transparent (or slightly yellowish) clear solu-tion after heating at 250   C. Meanwhile 100 mg(1.26 mmol) of Se is dissolved in 10 ml of liquid paraffinand 0.5 ml (2.0 mmol) tributhylphosphine (TBP) at150–200   C using reflux condenser in argon atmosphere.This also becomes transparent clear solution afterheating at 150–200   C. A portion of 2.5 ml of the hotTBP-Se solution is fast-injected into the hot-matrix of CdO, paraffin and stearic acid. The mixing makes thetransparent solution yellow which subsequently changesto red. Samples are taken from the reaction mixture atgiven time intervals for evaluation of the particle growth[6, 7]. In another experiment, the reaction mixture is immediately cooled down for isolation of CdSe nano-particles at a desired size.In the comparative experiments, TOPO (15 g,38.8 mmol) is used as a matrix solvent instead of liquidparaffin. It is found that cadmium acetate Cd(OAc) 2  canalso serve as a cadmium precursor in non-coordinatingsolvents such as liquid paraffin. In this case, 100 mg(0.37 mmol) of Cd(OAc) 2  dihydrate are used instead of CdO.The CdSe nanoparticles are purified from the matrixby the following way. A portion of 5 g of the cooledmatrix is dissolved in 100 ml of   n -hexane using anultrasonic bath for 5–10 min. The solution is centrifugedfor 5–10 min at 4,000 rpm with a rotor (10 cm in ra-dius). The sediment is washed with  n -hexane twice forbetter purification. The fluorescence of the particles re-mains the same when dispersed in different nonpolarsolvents:  n -hexane,  n -heptane, paraffin, toluene, etc. (atequal particle concentrations). For better purification,an extraction of the above dispersions with methanolcan be made.The growth kinetics of the nanoparticles is traced bymeasuring the absorbance and the fluorescence of thesamples, collected from the hot-matrix at time intervals15, 30, 60, 180, 300, and 480 s after the beginning of growth. A portion of 0.4 ml of each sample is diluted in2.5 ml of   n -hexane and 2–3 drops of TBP are added.Each sample is well agitated using a 15 kHz ultrasonicbath for about 5 min. After these procedures theabsorbance and fluorescence spectra are measured.Figure 1a shows the absorbance spectra at given timeintervals. The temporal evolution of the PL-maximum isshown in Fig. 1b. In both cases their peaks shift to largerwavelengths with the time implying on the growth of nanoparticles [6, 7]. CdS nanoparticles are synthesized and evaluated in asimilar way. In this case sulfur is used instead of sele-nium. The solution of tributylphosphine sulfide (TBP-S)is made by dissolving of 210 mg sulfur into 50 ml of liquid paraffin and 2.5 ml of TBP. This TBP-S is furtherused instead of TBP-Se in the synthesis. Figure 2 showsthe respective absorbance and PL-spectra of two sam-ples taken from the matrix at time intervals 35 and 55 safter the beginning of nanoparticle growth. Similarly theshifts toward larger wavelengths are observed, whichindicates nanoparticle growth. It is found that instead of TBP-S, a solution of sulfur in liquid paraffin (4.2 mg/ml)could also be applied as a precursor.In Fig. 3 are shown TEM images of the stearic acidcapped QDs. The rather regular spacing between theinorganic QDs is probably an evidence of the presence of a capping layer around them. Fig. 1  Temporal evolution of the absorbance ( a ) and photolumi-nescence ( b ) spectra of CdSe nanocrystals, prepared at 250   C inliquid paraffin matrix using CdO precursor. The time afterbeginning of nanocrystals growth (in seconds) is shown with numbers on the lines . Spectra are taken in hexane solutions at aroom temperature230  Discussion In order to compare the two matrices, coordinating(TOPO) and non-coordinating (liquid paraffin), twoexperiments have been made. The reaction temperature,the precursor concentrations and the concentration of stearic acid are the same except use of TOPO instead of use of liquid paraffin. The nanoparticle size throughoutthe growth is calculated from the optical spectra inFig. 1a using the equations derived in [7]. The calculated radii of nanoparticles are plotted in Fig. 4 in the case of CdSe. As seen, they are composed of two growthkinetics—fast and slow, which is very similar to thatobserved with CdSe nanoparticles in pure TOPO matrix[7]. The fast increase of the radius at the beginning of nanoparticle growth is due to a reaction-driven processduring which the precursor from the nanocrystal vicinityadsorbs on its surface crossing the capping layer of surfactant molecules there. This process exhausts thesubsurface layer, adjacent to the nanoparticle surface,from material, which leads to sharpening of the nano-particle size distribution known as focusing [6]. Thesubsequent slow process at large times is a diffusion-limited growth, when the precursor moves from distinctparts of the suspension to the subsurface layer. Quan-titatively, this kinetics is described by the set of equa-tions, outlined in [7]—see Eq. 33 for the fast growth andEq. 34 for the slow one. Table 1 summarizes the valuesof fitting parameters used in the calculations for the twomatrices.Further growth of the nanocrystals can be due toOstvald ripening—this is the ‘‘defocusing of size distri-bution’’ and should be avoided for the formation of relatively monodisperse colloidal nanocrystals.The results in Fig. 4 and Table 1 show that the nucleation and growth processes are slower when TOPOis used. Also, smaller nanocrystals are obtained in theTOPO matrix. In both cases, the Cd-monomer is Cd-stearate that is formed in situ by heating the CdO andstearic acid. However, the fatty acid is found to beabsolutely necessary for the growth of nanocrystals inthe non-coordinating solvent (paraffin). To prove thisfact more experiments are made in liquid paraffin and Fig. 2  Temporal evolution of the absorbance ( a ) and photolumi-nescence ( b ) spectra of CdS nanocrystals, prepared by using TBP-S/paraffin precursor at 250   C. The time after beginning of nanocrystals growth (in seconds) is shown with  numbers on the lines Fig. 3  High-resolution TEM images of ( a ) CdS (mean diameter2.7 nm) and ( b ) CdSe (mean diameter 3.0 nm) nanocrystalsprepared in paraffin matrix Fig. 4  Comparison between the temporal evolution of the meanCdSe nanocrystals radius in paraffin matrix and in TOPO matrix at250   C. The  solid lines  are theoretically predicted. The  error bars are deviations from the mean value of two experiments231  TOPO. In two of them cadmium acetate is used insteadof CdO. First, Cd(OAc) 2  is heated together with stearicacid in paraffin to remove the crystal water and to formCd-stearate. The addition of TBP-Se solution producesCdSe nanocrystals with almost similar growth ratecompared to that when CdO is used for the in situ for-mation of Cd-stearate. Second, when we use Cd(OAc) 2 at the same conditions but without stearic acid, nano-particles are obtained only in TOPO solvent, pure orwith paraffin (their growth is very fast and significantlylarge nanocrystals are created). On the contrary, CdSebulk material (large powder) is formed in pure liquidparaffin solvent (without stearic acid and withoutTOPO). The possibility to form nanocrystalline CdSe inpure TOPO by using Cd(OAc) 2  has been previously re-ported [8]. Use of CdCl 2  and CdSO 4  instead of Cd(OAc) 2  also results in the formation of bulk-sizedCdSe in the tested conditions of paraffin with or withoutstearic acid. Obviously, these salts do not form Cd-car-boxylate.According to these results, the following suggestionscan be made. In the Cd-carboxylate/TOPO system bothTOPO and carboxylate ions have played a significantrole to stabilize the nanocrystals (where carboxylatemeans stearate or acetate). This conclusion is proven bythe formation of bulk CdSe in the absence of carbox-ylate ions (when CdCl 2  and CdSO 4  are used) or in theabsence of TOPO (when paraffin is used). However, thepresence of carboxylate ions is much more importantthan the free acid, since CdCl 2  and CdSO 4  could notform nanocrystals in the absence of carboxylate at alltested conditions. The chain of the carboxylate ionshould be longer for better stabilization of the nano-crystals, e.g., stearate is more preferable than acetate. Inthe case of small carboxylate ion (acetate), TOPO playsa more significant role in the nanocrystal growth—theabsence of TOPO in the Cd(OAc) 2 /paraffin system leadsto bulk CdSe. Conclusions A composite paraffin matrix, composed of thermoresis-tant organics and surface-active compound, can suc-cessfully replace TOPO in the formation of high qualityA II B VI nanoparticles. The composite matrix has twofunctions: the container of nanospecies and the coordi-nating agent providing the formation and growth of nanoparticles. Each function comes from liquid paraffin(bulk medium) and stearic acid (surfactant), respec-tively. The stearic acid might form in the liquid paraffina sort of aggregates like complexes or nano-micelles withthe hydrocarbon tails facing the paraffin. However, suchcomplexes or nano-micelles are thermally generated ordestroyed and these processes should have fluctuativecharacter. These nano-micelles, if existing, serve as thedensity fluctuations or nano-cavities in the matrix forthe nucleation of the semiconductor nanocrystals. Thelonger the carboxylate chain, the better the stabilizationof the nanocrystal in the nonpolar medium. Also, thelonger carboxylate chain is a reason for the slower dif-fusion of the Cd-monomer in the solution and troughthe subsurface layer around a nanoparticle. This slowsdown the reaction rate and provides better control overthe nanocrystal growth. A combination of the paraffinsolvent with TOPO may also provide fine control of thenanocrystals, depending on the TOPO concentration inthe solution. For example, at higher TOPO to paraffinratio the crystal growth rate will be slower and viceversa.The paraffin composite matrix (liquid paraffin andstearic acid) is a low-cost and environmentally accept-able compound. This opens the way for manufacturingof semiconductor nanoparticles by the hot-matrixpyrolysis of precursors in the large scale. Table 1  Parameters of the numerical fits of data in Fig. 4 R 0  (nm)  d  (nm)  R r  (nm)  R d  (nm)  s r  (s)  s d  (s)Liquid paraffin 0.80 1.9 1.60 1.75 45 2,000TOPO 0.75 1.9 1.44 1.55 55 1,000 R 0  is the radius of initial nuclei preceding the abundant nanopar-ticles;  d  is the thickness of subsurface layer surrounding a nano-particle;  R r  and  R d  are the limiting radii of fast and slow growth,respectively;  s r  and  s d  are the characteristic time constants of thefast and slow process, respectively References 1. Murray C, Norris D, Bawendi MJ (1993)Am Chem Soc 115:87062. Bowen Katari JE, Colvin VL, AlivisatosAP (1994) J Phys Chem 98:41093. Peng ZA, Peng X (2001) J Am Chem Soc123:1834. Pileni MP (1993) J Phys Chem 97:69615. William W, Peng X (2002) Angew ChemInt Ed 41:23686. Peng X, Wickham J, Alivisatos AP(1998) J Am Chem Soc 120:53437. Dushkin C, Saita S, Yoshie K, Yamag-uchi Y (2000) Adv Colloid Interface Sci88:378. Qu L, Peng Z, Peng X (2001) Nanoletters1:333232
Related Search
Advertisements
Similar documents
View more...
Advertisements
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks