DOI: 10.1021/cg901490w

2010, Vol. 10 1813–1822

)

Surface Activity of Highly Hydrophobic Surfactants and Platelike PbSe and CuSe Nanoparticles Mandeep Singh Bakshi,*,† Pankaj Thakur,§ Poonam Khullar, Gurinder Kaur,‡ and Tarlok Singh Banipal§ †

Department of Chemistry, Mount Saint Vincent University, Halifax, Nova Scotia, B3M 2J6 Canada, Nanotechnology Research Laboratory, College of North Atlantic, Labrador City, Newfoundland A2V 2K7 Canada, §Department of Applied Chemistry, Guru Nanak Dev University, Amritsar 143005, Punjab, India, and Department of Chemistry, B.B.K. D.A.V. College for Women, Amritsar 143005, Punjab, India )

‡

Received November 29, 2009; Revised Manuscript Received January 7, 2010

ABSTRACT: Lead selenide (PbSe) and copper selenide (CuSe) nanoparticles were synthesized in aqueous phase at a relatively mild temperature (85
C) in the presence of various cationic Gemini surfactants (12-2-12, 12-0-12, and 16-2-16) as capping/stabilizing agents. All nanoparticles exhibited clear core-shell (surfactant) morphologies. PbSe reactions produced predominantly platelike cubic morphologies along with long Se nanorods (NRs) as a reaction byproduct. CuSe particles were polyhedral thin plates with perforations. High resolution transmission electron microscopy (HRTEM), field emission scanning electron microscopy (FESEM), and X-ray diffraction (XRD) measurements were used to characterize the shape and structure of the particles. HRTEM allowed us to measure the precise thickness of the surfactant shell around each nanocrystal (NC) which was in excellent agreement with the length of the surfactant hydrocarbon tail. Infrared spectroscopic (FT-IR) studies suggested a strong affinity of cationic surfactant for NC surface which was the driving force for the monolayer formation in the form of a shell. Energy dispersive X-ray spectroscopic (EDS) analysis demonstrated that PbSe and CuSe particles were always in 1:1 stoichiometry, and Se NRs were made up of only pure Se and no Pb contents were observed. Stronger interfacial adsorption of a surfactant with greater hydrophobicity controlled the morphology to produce platelike geometries. The size of PbSe and CuSe particles increased while the thickness decreased as the hydrophobicity of the surfactant increased in the order of 12-212 < 12-0-12 < 16-2-16.

*To whom correspondence should be addressed. E-mail: ms_bakshi@ yahoo.com.

synthesis of PbSe16 and CuSe17 NCs. Surfactant-assisted pathways are the simplest in view of the environmental concerns. They are reproducible and best suited for their high yield. Among the ionic surfactants, we have already used cationic Gemini surfactants as capping/stabilizing agents for the shape-controlled synthesis of PbS NCs.18 They show remarkable control over crystal growth at nanoscale and produce well-defined geometries. Shape-controlled morphologies are the usual building blocks of nanodevices, and their synthesis in aqueous phase has advantages both from economical and environmental perspectives. In the present work, we have used Gemini surfactants to synthesize PbSe and CuSe NCs at a relatively low temperature of 85
C by using acetates of lead and copper as the metal source, and sodium selenite (Na2SeO3) as the Se source, in an aqueous medium. In the case of PbSe, apart from well-defined core-shell nanogeometries such as cubes and hexagons, we also obtained long Se nanorods of several hundred nanometers without any surfactant coating. On the other hand in the case of CuSe, predominantly spherical platelike geometries with a well-defined surfactant shell were obtained without the appearance of Se nanorods. The formation of platelike morphologies is the novelty of this synthesis because such morphologies generate an enormous surface area for the interfacial selective adsorption of strongly hydrophobic Gemini surfactant molecules. Such a selective adsorption allows the formation of platelike core-shell morphologies. This work reports the mechanistic aspects of a simple and relatively low temperature synthesis of such novel materials.

r 2010 American Chemical Society

Published on Web 01/20/2010

Introduction Components of the next generation technologies such as the semiconductor nanocrystals (NCs) are going to be materials with nanometric dimensions1 with many useful applications in lasers,2 photovoltaics,3 light emitting diodes,4 and biological assays.5 Lead chalogenides are of extensive interest among the semiconductor NCs6 due to their narrow band gap and large Bohr’s radius. Lead selenide (PbSe) with the narrowest band gap is synthesized by solvothermal7 and microwave-assisted methods.8 Copper selenide (CuSe), another semiconducting material, is used as a superionic material9 as well as an optical filter in solar cells.10 Several methods of synthesis of CuSe NCs include thermolysis of Cu and Se powder mixtures at 400-470
C in the presence of flowing argon,11 solvothermal process,12 mechanical alloying of Se and Cu in a high energy ball mill,13 and in liquid ammonia.14 Chalcogenide NCs exhibit unique size and shape dependent physicochemical characteristics,15 which are the outcome of low dimensional quantum confinement effects and have a wide range of applications as mentioned above. They have also been effectively employed as building blocks for two- and three-dimentional arrays for the development of metamaterials with collective electronic and optical properties.15g Conventional surfactants have been widely used for the shape-controlled synthesis of noble metal nanoparticles of gold and silver, but few reports are related to their use in the

pubs.acs.org/crystal

1814

Crystal Growth & Design, Vol. 10, No. 4, 2010

Bakshi et al.

Experimental Section Materials. Lead acetate, copper acetate, hydrazine, and sodium selenite all 99% were purchased from Aldrich. Cationic Gemini surfactants (Scheme 1) didodecyl dimethylammonium bromide (12-0-12, critical micelle concentration, cmc = 0.2 mM), dimethylene bis(dodecyldimethyl ammonium bromide) (12-2-12, cmc = 0.84 mM), and dimethylene bis(hexadecyldimethyl ammonium bromide) (16-2-16, cmc = 0.02 mM) were synthesized as reported in the literature.19 Double-distilled water was used for all preparations. Synthesis of Surfactant Core-Shell PbSe and CuSe Nanoparticles. PbSe NCs were obtained through a reaction between lead acetate and sodium selenite in the presence of aqueous hydrazine. In a typical reaction, 10 mL of aqueous surfactant (12-0-12/ 12-2-12/16-2-16) solution of concentration 4, 8, or 16 mM containing 2.5% hydrazine was taken in a round-bottom glass flask. A fresh hydrazine should be used. Then, 1.25 mM of lead acetate was added to it. This was followed by the addition of 0.125 mL of 1.25 mM aqueous sodium selenite solution under constant stirring. After all the components were mixed at room temperature, the reaction mixture was kept in a water thermostat bath (Julabo F 25) at a fixed temperature of 85 ( 1
C for 48 h. The color of the solution changed from slightly turbid white due to the low solubility of surfactant initially to clear dark brown and finally it turned black within 4 h and remained the same until 48 h. A colloidal black solution thus obtained indicates the presence of PbSe NCs. The

Scheme 1

purified sample was collected by centrifugation at 10 000 rpm for 5 min after washing it 2-3 times with distilled water. Recently, Kotov15h-15j et al. have also synthesized similar composite silica and semiconductor morphologies by using oxygen-free synthesis. But the present synthesis is simple and straightforward where aqueous Gemini surfactant has been directly used as a capping layer in order to form the shell around semiconductor NCs. Stable aqueous micellar phase is essential to achieve such morphologies. The same procedure was adopted for the preparation of CuSe nanoparticles by using copper acetate in place of lead acetate. The color of the solution containing CuSe colloidal suspension changed from light black to dark black within 2 h and remained the same for 48 h. All PbSe and CuSe samples along with their reaction ingredients are listed in Table 1. Methods. The shape and size of nanoparticles were characterized by transmission electron microscopy (TEM). Size distribution histograms were drawn manually by using appropriate TEM images. The samples were prepared by mounting a drop of a solution on a carbon-coated Cu grid and allowing it to dry in the air. They were observed with the help of a Philips CM10 transmission electron microscope operating at 100 kV. High resolution TEM (HRTEM) and high-angle annular dark field (HAADF) images were determined with the help of CM20. Energy dispersive X-ray spectroscopic (EDS) analysis was also performed with the same instrument. Scanning electron microscopy (SEM) measurements were carried out by using Hitachi S-4800 field emission scanning electron microscope. A few drops of a diluted suspension of each sample were placed on a piece of silicon wafer and dried in the air, and then observed at accelerating voltage of 5 kV. X-ray diffraction (XRD) patterns were recorded using a Bruker-AXS D8-GADDS with Tsec = 480. Samples were prepared on glass slides by putting a concentrated drop of aqueous sample on them and then drying in a vacuum desiccator. The UV-VIS spectra of some of the samples were recorded in the range of 200-900 nm by using a UV spectrophotometer (Perkin-Elmer Lambda 25). FT-IR spectra were taken by using a FT-IR spectrometer (Shimadzu) in the range of 4000-400 cm-1. A few drops of a concentrated aqueous colloidal suspension were placed in the center of a clean silicon wafer equal to the size of a spectrophotometer window. The sample was then dried by keeping it in the vacuum oven and then loading it onto the spectrophotometer window. Each spectrum was measured in transmission mode with 256 scans and 4 cm-1 resolution. The polarity of the cationic Gemini surfactant capped nanoparticles was determined by the gel electrophoresis by using tris-HCl buffer as a gel running medium with pH = 7. For this purpose, 1% of aqueous agrose solution was first microwave boiled and left in the gel plate to harden. Then, 20 μL of colloidal aqueous suspension was loaded in each gel well and a direct voltage of 90 V was applied for 30 min to observe the movement of NCs. No staining agent was used because the NC suspension in each case was black colored.

Results and Discussion PbSe Nanocrystals and Se Rods. As mentioned in the Experimental Section, PbSe NCs were synthesized by using 12-2-12 (PbSe1), 12-0-12 (PbSe2), and 16-2-16 (PbSe3) Table 1. Various Constituents of Samples of PbSe and CuSe, and the Shape and Structure of their Nanoparticles and Se Nanorods samples

[Pb(CH3COO)2]/[Na2SeO3]

[surfactants]/16 mM

PbSe NCs (shape and size)

PbSe1

1

12-2-12

PbSe2

1

12-0-12

PbSe3

1

16-2-16

cubic, hexagonal 26.3 ( 8.4 nm cubic 39.1 ( 9.1 nm cubic 69.0 ( 26.3 nm

samples

[Cu(CH3COO)2]/[Na2SeO3]

[surfactants]/16 mM

CuSe NCs (shape and size)

CuSe1

1

12-2-12

CuSe2

1

12-0-12

CuSe3

1

16-2-16

nanoplates 94.7 ( 23.8 nm nanoplates 106 ( 31 nm nanoplates 116 ( 33 nm

Se rods needles several hundred nm needles several hundred nm needles several hundred nm Se rods no needles no needles no needles

Article

Crystal Growth & Design, Vol. 10, No. 4, 2010

1815

Figure 1. Panel (a) shows several PbSe core-shell particles of sample PbSe1 (Table 1) synthesized in the presence of 12-2-12 = 16 mM. Three insets represent selected area diffraction image, magnified view of a few core-shall particles clearly demonstrating the presence of a surfactant layer around each particle, and HAADF image of a few such particles. (b) A lattice resolved image of one cubic-shaped particle. (c) EDX spectrum with clear peaks due to Pb and Se. (d) XRD patterns of particles of (a) with sharp and narrow peaks indexed to the PbSe cubic phase. (e) SEM of long Se nanorods and (f) shows its closeup of a dark field image with corresponding EDX (g) with peaks due to only Se, and no Pb contents were observed.

Gemini surfactants (Table 1). The electron micrographs of clear crystalline core-shell PbSe NCs of sample PbSe1 are shown in Figure 1a. The corresponding size distribution (26.3 ( 8.4 nm) histogram is shown in Supporting Information, Figure S1. They are mainly cubic or hexagonal in shape. Each NC is completely covered with a continuous film of

surfactant molecules which is quite clear in both bright and dark field images (see insets). Figure 1b shows a lattice resolved image of single PbSe NC of cubic geometry oriented along the Æ110æ zone. One can easily differentiate it from the hexagonal NC shown in the dark field image (inset, Figure 1a). Both images indicate the single crystal nature

1816

Crystal Growth & Design, Vol. 10, No. 4, 2010

Bakshi et al.

Figure 2. (a) Platelike core-shell morphologies of predominantly cubic geometries of PbSe particles with a clear surfactant coating of sample PbSe2 (Table 1) synthesized in the presence of 12-0-12 = 16 mM. Inset shows a selected area diffraction image. (b) Lattice resolved image of one particle showing the presence of ≈1.5 nm surfactant coating. Inset demonstrates the gel electrophoresis. (c) XRD patterns of particles of (a) with sharp and narrow peaks indexed to PbSe cubic phase. (d) Lattice resolved image of a single platelike particle showing the growth direction along the {100} crystal planes, while a dark field image (inset) further confirms the platelike geometry. (e) SEM image of a single long Se nanorod and yellow dotted circle represents the area used for EDX analysis. TEM image of similar several rods is shown in the inset. (f) EDX spectrum clear demonstrates peaks only due to Se, and no Pb contents were observed.

Article

Crystal Growth & Design, Vol. 10, No. 4, 2010

1817

Figure 3. (a) Several polyhedral platelike geometries of CuSe nanoplates of sample CuSe3 (Table 1) synthesized in the presence of 16-2-16 = 16 mM. The yellow dotted circles show some of the plates with clear perforations. Selected area diffraction image is shown in inset. (b) SEM image of the same sample confirming the presence of a platelike morphology of these particles. (c) EDX spectrum of this sample with 1:1 atomic percent of Cu and Se. (d) XRD patterns of particles of (a) with sharp and narrow peaks indexed to CuSe hexagonal phase. (e) Lattice resolved image of a portion of a single plate with bright shell of ≈2.5 nm surfactant coating. Inset shows a further low resolution image pointing to a continuous surfactant layer. (f) A plot of UV-visible absorbance of colloidal aqueous suspension of nanoplates.

of the particles. The close inspection of several such particles suggested that most of the NCs are cubic geometries. The

hexagonal particles in fact lead to the cubic geometries when the growth rate is higher in the Æ111æ direction than in <110/100>

1818

Crystal Growth & Design, Vol. 10, No. 4, 2010

directions.16c Figure 1c scans its EDS spectrum. Emissions from Pb and Se are evident with 10 and 8 atomic %, respectively, relative to C and O. XRD patterns (Figure 1d) show sharp and narrow peaks corresponding to cubic phase of PbSe (JCPDF No. 060354). When the same sample is carefully purified at 4000 rpm for 5 min, only large particle are collected while nanometric NCs remain suspended in the solution. A low resolution field emission scanning electron microscopy (FESEM) image (Figure 1e) of this sample demonstrates the presence of many long nanorods (NRs) of several hundred nanometers. A close inspection indicates that they have needlelike morphology (Figure 1f). Surprisingly, the EDS analysis (Figure 1g) shows only peaks due to Si (surface) and Se in 24 atomic % contents relative to C and O, which means that the needles are made up of only Se and no Pb contents are observed. Figure 2 on the other hand shows various images of PbSe NCs and Se NRs of sample PbSe2 synthesized in the presence of 12-0-12. This sample contains mostly cubic crystalline geometries of 39.1 ( 9.1 nm in size (Figure S2, Supporting Information) with a fine shell of 12-0-12 around each NC. A lattice resolved image is shown in Figure 2b. One can find a surfactant monolayer of ≈1.5 nm thickness. Figure 1a also gives almost the same value of thickness for the 12-2-12 monolayer, because both 12-0-12 and 12-2-12 have identical tail lengths, that is, C12. This value is in good agreement with the radius of the dodecyltrimethyl ammonium bromide (DTAB) micelle with a C12 carbon tail length, that is, 1.8 nm.20 A slightly larger latter value can be attributed to the hydrated state of the micelle in comparison to the dried monolayer on the surface of NC. The presence of a surfactant monolayer around each NC was further confirmed from the gel electrophoresis (Figure 2b, inset). All samples show displacement toward the negatively charged electrode indicating the fact that PbSe NCs are positively charged due to the presence of monolayer of cationic surfactants. XRD patterns (Figure 2c) fully characterize the cubic phase of PbSe. There is a clear difference in the intensities of (200) and (220) crystal planes from the corresponding peaks of Figure 1d. In Figure 1d, the intensity of (220) is much higher in comparison to that of (200), while the reverse is observed for Figure 2c. Though both kinds of particles predominantly show cubic geometries, the higher intensity of the (200) peak in the case of PbSe2 suggests a greater number of NCs are bound with {100} crystal planes of the PbSe cubic phase. Such crystal planes are clearly visible in Figure 2d oriented along the Æ100æ zone. The cubic geometries are platelike morphologies which is evident from the HAADF image of single NC (inset). A careful purification procedure again showed the presence of long Se needles of several hundred nanometers (Figure 2e). An EDS spectrum of a single needle (Figure 1f) clearly demonstrates the presence of only Se in 43 atomic % relative to the other elements further confirming the fact that long needles are indeed of only Se. Furthermore, when the same reaction is carried out in the presence of 16-2-16 (sample PbSe3, Table 1), we again got cubic geometries of somewhat larger dimensions along with long nanorods (Figure S3, Supporting Information). Essentially similar results were obtained at other two lower concentrations (4 and 8 mM) of 12-2-12, 12-0-12, and 16-2-16 (not shown). CuSe Nanocrystals. Under similar reaction conditions, we obtained CuSe nanoparticles but without the presence of any long Se NRs. Figure 3 shows the CuSe particles of sample

Bakshi et al.

CuSe3 (Table 1) synthesized in the presence of 16-2-16. All particles are single crystal and thin nanoplates (NPs) of polyhedral geometries with size of 116 ( 33 nm (Figure S4, Supporting Information). Corresponding FESEM image (Figure 3b) further confirms the platelike geometries. The thickness of each NP is ∼5-8 nm, which is approximately 1/18 of the overall size. EDS spectrum (Figure 3c) of this sample confirms the presence of Cu and Se in 1:1 atomic % because their stoichiometric amounts in the reaction are also in a 1:1 ratio (Table 1). XRD patterns (Figure 3d) further confirm the presence of a hexagonal phase of CuSe which is fully supported by the diffraction image shown as an inset in Figure 3a. All the rings can be indexed to the reflection of hexagonal CuSe crystals. A HRTEM lattice resolved image of a single NP (Figure 3e) shows a clear bright shell of surfactant monolayer (inset). Close inspection of the shell indicates that it is ∼2.5 nm thick, which is very much in agreement with the radius of the hexadecyltrimethyl ammonium bromide (HTAB) micelle,20 that is, 2.3 nm with an identical chain length of C16. Apart from this, the well-defined geometries of NPs also produce absorbance around 460 nm in their colloidal suspensions (Figure 3f). Usually, other calcogenides of Cu such as CuS show much broader absorbances around 600 nm.21 But Pileni et al.22 on the basis of a systematic study on Cu nanoparticles demonstrated that small particles of various shapes including nanodisks show a broad absorbance around 600 nm which blue shifts as the size increases. We attribute the absorbance at lower wavelength to well-defined predominantly monodisperse NPs. Interestingly, almost all NPs show perforations (see dotted circled NPs in Figure 3a) which are clearly visible in the HAADF image shown in Figure 4a. EDS line spectrum was performed to confirm their presence (Figure 4b) as well as the composition of single NP with respect to the amounts of Cu and Se. Homogenous mixing between Cu and Se to form 1:1 CuSe is evident from the elemental mapping presented in Figure 4c,d. However, we do not find any perforated NPs in the presence of 12-2-12 (sample CuSe1, Figure 4e) and 12-0-12 (sample CuSe2, Figure 4f) under similar reaction conditions. The average sizes of NPs obtained in the presence of 12-2-12 and 12-0-12 are 94.7 ( 23.8 nm (Figure S5, Supporting Information) and 106 ( 31 nm (Figure S6, Supporting Information), respectively. FTIR Studies. FTIR spectral studies of PbSe NCs were performed to determine the interactions of capping surfactant molecules with PbSe surfaces. The FTIR spectra of pure surfactants, that is, 12-2-12, 12-0-12, and 16-2-16, have been compared with surfactant capped PbSe NCs at a surfactant concentration of 16 mM (Figure 5). The IR spectrum of pure 12-2-12 shows two scissoring modes of vibrations (δs(C-H)) at 1473 and 1465 cm-1, which shift to 1470 and 1459 cm-1 in the case of 12-2-12 capped PbSe NCs. For 12-0-12, only one scissoring mode of vibration (δs(C-H)) at 1470 cm-1 shifts to 1466 cm-1. A significant shift in νasym(Nþ-CH3) mode from 1436 to 1455 cm-1 can be attributed to the interhead groups repulsions which might arise due to the formation of compact monolayer in view of strong hydrophobic interactions among the twin tails of adjoining surfactant molecules. In addition, the number of ν(C-Nþ) modes (i.e., 1160, 1056, and 986 cm-1) is more in case of 12-2-12 as compared to that of 12-0-12 (i.e., 978 and 947 cm-1) which demonstrates that some of the 12-0-12 headgroup stretching modes are absent in view

Article

Crystal Growth & Design, Vol. 10, No. 4, 2010

1819

Figure 4. (a) HAADF image of a single nanoplate showing dark perforations. (b) EDX line spectrum confirming the presence of perforations as well as 1:1 composition of Cu (c) and Se (d). Several nanoplates without perforations synthesized in the presence of 12-2-12 = 16 mM (e) and 12-0-12 = 16 mM (f).

of stronger interactions between 12-0-12 and PbSe surface. Though 12-2-12 and 12-0-12 have identical tetralkylammonium head groups, a stronger hydrophobicity of 12-0-12 could be responsible for its stronger interactions at the PbSe surface. This is because the presence of two methylene groups as spacer in the headgroup region of 12-2-12 induces steric hindrances in a compact alignment of twin surfactant hydrophobic tails, which is obviously absent in the case of 12-0-12. Likewise for 16-2-16, one

νasym(Nþ-CH3) mode at 1492 cm-1 shifts to 1511 cm-1 along with a slight shift in the scissoring mode of vibration (δs(C-H)) from 1476 to 1478 cm-1. Similar information can be drawn from FTIR studies on surfactant capped CuSe NPs (Table 2). Reaction Mechanism (Formation of Se NRs). The occurrence of Se NRs along with PbSe NCs seems to be due to a high concentration of hydrazine (2.5%) used in the present reactions. Lead acetate and sodium selenite are expected to

1820

Crystal Growth & Design, Vol. 10, No. 4, 2010

Bakshi et al.

dissociate into their respective ions as soon as they are dissolved in water (eqs 1 and 2, respectively). Hydrazine will convert selenite ions into selenium ions by following a redox reaction 3. Once selenium ions are produced, they will undergo neutralization with lead ions to generate lead selenide in the presence of hydrazine 4. Small nucleating centers will achieve colloidal stability in the presence of surfactants which will act as capping/stabilizing agents. Further growth should follow the autocatalytic process. PbðCH3 COOÞ2 ðsÞ f Pb2þ ðaqÞ þ 2CH3 COO - ðaqÞ Na2 SeO3 ðsÞ f 2Naþ ðaqÞ þ SeO3 - ðaqÞ

N2 H4 ðaqÞ þ SeO3 2 - ðaqÞ f SeðsÞ þ N2 ðgÞ þ 2OH - ðaqÞ þ ð5Þ

H2 OðlÞ

ð1Þ ð2Þ

3N2 H4 ðaqÞ þ SeO3 2 - ðaqÞ f 2Se2 - ðaqÞ þ 3N2 ðgÞ þ 6H2 OðlÞ

ð3Þ Pb2þ ðaqÞ þ Se2 - ðaqÞ f PbSeðsÞ

However, hydrazine a mild reducing agent can easily reduce Se(IV) to Se0(V) with þ0.74 V standard reduction potential, whereas this is not so for Pb2þ/Pb0 with -0.13 V. The origin of this might come from

ð4Þ

Figure 5. FTIR spectra of pure 12-2-12, 12-0-12, 16-2-16, and that of PbSe NCs prepared in the presence of these surfactants. See text for details.

an excessive amount of hydrazine used herein. Being bifunctional, hydrazine in excessive amounts will obviously shield Pb2þ ions to interact with Se2- ions through water-soluble adduct formation 6. Under such a situation, only reaction 5 will proceed and the free SeO32- ions will continue the autocatalytic reaction. However, this situation may not be occurring in the case of CuSe NPs because no Se NRs were observed. Cu2þ ion on the other hand is a stronger oxidizing agent than Pb2þ ion, and hence it will have a greater affinity for Se2- ion in order to form CuSe in the presence of hydrazine. Effect of Surfactant Hydrophobicity on the Morphology of PbSe and CuSe Particles. 12-0-12 is at least four times more hydrophobic than 12-2-12 on the critical micelle concentration scale23 due to the lack of one ammonium group and two methylene group spacer that reduces a considerable hydration in the headgroup region of aggregated surfactant. We do not observe any major morphology change in PbSe NCs from 12-2-12 to 12-0-12, but a higher intensity of (200) diffraction peak in the presence of 12-0-12 suggests that more NCs are bound with {100} planes. A stronger interfacial adsorption of 12-0-12 in comparison to that of 12-2-12 is expected to provide a better protection to {100} crystal planes from participating in further nucleation. Likewise, 16-2-16 contains two hydrocarbon chains of C16 with four methylene groups extra in comparison to that of 12-2-12 and 12-0-12. The four extra methylene groups are expected to induce even greater hydrophobicity which provides a better capping ability and a stronger liquid-solid interfacial adsorption. 16-2-16 is considered to be at least 40 and 10 times more hydrophobic than 12-2-12 and 12-0-12, respectively.23 Again, we do not observe any significant difference between the morphologies of CuSe NPs synthesized in the presence of various surfactants, but the XRD patterns of CuSe NPs prepared in the presence of 12-0-12 and 12-2-12 clearly show less predominant growth at (006) in comparison to (102) and (110) (Figure S7, Supporting Information). It means that {100} crystal planes of CuSe NPs are relatively less protected by 12-2-12/12-0-12 in comparison to 16-2-16 molecules.

Table 2. Mode Assignments of Various Surfactants in the Absence and Presence of PbSe/CuSe NCsa peaks assignment 12-2-12 12-2-12-PbSe 12-2-12-CuSe 12-0-12 12-0-12-PbSe 12-0-12-CuSe 16-2-16 16-2-16-PbSe 16-2-16-CuSe δs (C-H) νasym (Nþ-CH3) ν (C-Nþ)

a

1473 1465

1470 1459

1480 1460

1160 1056 986

1156 1050 979

1158 1049 982

1470

1466

1464

1476

1478

1481

1436 978 947

1455 970

1456 969

1492 982

1511

1516 975

ν = stretching, sym = symmetric, asym = anti-symmetric, δs = methylene scissoring.

Article

Mechanism of Surfactant Adsorption. TEM images demonstrate that the thickness of the surfactant shell is always equivalent to a single layer of surfactant monomers. Since the colloidal particles are stabilized in aqueous phase, cationic Gemini surfactant head groups should be in contact with water in order to maintain the colloidal stability, while hydrophobic tails should be in contact with the surface of PbSe NCs and CuSe NPs. This is only possible if both kinds of particles have zero dipole moment. Cho et al.16c have recently explained a systematic relation between the lattice structure and dipole moment of nanoparticles. A nonzero dipole moment of chalcogenides allows them to interact with each other and that subsequently leads to the formation of superstructures. However, we do not observe any fusion among the PbSe NCs and CuSe NPs. The particles are singly dispersed or associated with each other through the surfactant coating which indicates they have zero dipole moment. The zero dipole moment is primarily associated with the lattice structure. The {100} crystal facets are mainly made up of both Pb/Cu and Se atoms, and hence they are without any charge polarization (due to a difference in their electronegativities). Contrary to this, {111} crystal planes are predominantly occupied by either Pb/Cu or Se atoms and that leads to an overall charge separation and establishes a dipole. The NCs shown in Figures 1a and 2a are of cubic geometry bound with {100} and {110} crystal planes and are considered to have zero dipole moment.16c In addition, the zero dipole moment provides nonpolar properties to these planes which will in turn act as suitable platforms for surfactant adsorption. Thus, nonpolar interactions will attract the surfactant hydrophobic tails to have a monolayer formation while head groups will be in contact with the aqueous phase. This will protect the {100} facets from further growth and hence will direct the growth at {110} or {111} crystal planes. Since the {111} facets possess higher surface energy, they are eliminated during the growth in the presence of surfactant. This mechanism has been observed for even PbSe nanowire formation.16c In the present work, replacing 12-2-12 with even stronger surfactant 12-0-12 dramatically reduces the intensity of (220) in comparison to that of (100) (Figure 2c) indicating the fact that predominantly more NCs are bound with {100} facets. A more tightly packed surfactant layer is more effective in reducing the participation of {100} planes in the crystal growth. It may produce more flat geometries (Figure 5) and consequently lead to an increase in the size with the increase in the hydrophobicity of the surfactant (Table 1). On the other hand, the formation of platelike geometries in the case of CuSe is again the consequence of an effective surfactant adsorption on {100} crystal planes. A hexagonal unit cell contains six atoms and is more tightly packed than a simple cubic phase of PbSe with four atoms per unit cell. Therefore, {100} crystal planes of CuSe possess more number of atoms than that of PbSe, which would provide the former with a greater surface energy than the latter to interact strongly with surfactant hydrophobic tails. Hence, 16-2-16 will provide better and tighter surface coverage compared to 12-2-12/12-0-12 and will result in even thinner NPs formation. The thinner NPs are more prone to crystal defects and might lead to surface perforations as we observed them in the presence of 16-2-16 only (Figure 4a). However, we do not observe similar geometries for Se, which is also a byproduct of PbSe reaction. Only long

Crystal Growth & Design, Vol. 10, No. 4, 2010

1821

one-dimensional (1D) NRs of several hundred nanometers are obtained. There are many reports of trigonal(t)-Se long rods in the literature,24 which is the most stable allotropic form of Se with semiconducting properties. Ma et al25 have reported the formation of t-Se nanotubes from nonionic surfactant micellar solutions. It has been proposed that nucleation occurs at the circumferential edges of hexagonally or trigonally faceted seeds, leading to the formation of t-Se nanotubes. A survey of all these studies24 simply leads to a conclusion that different kinds of stabilizing agents including nonionic and ionic surfactants, amino acids, and polymers only help the growing t-Se nucleating centers in achieving the colloidal stability rather than directing a crystal growth in a specific direction. Otherwise, we might have seen different morphologies in the presence of different stabilizing agents due a large difference in their interfacial properties. In other words, t-Se is a thermodynamically stable form of Se and hence the growth is favored only in the Æ100æ direction and that leads to the formation of mostly 1D nanostructures. Since we do not observe any clear surfactant capping on long Se NRs as observed in the case of PbSe/CuSe nanoparticles, it indicates that highly surface active Gemini surfactants (used in the present study) have the least affinity for t-Se surface. Otherwise, we could have obtained well-defined nanogeometries of t-Se as observed for PbSe and CuSe. The most probable reason for this might be due to the high electron affinity value of Se after halogens. Thus, Se will not have any significant interactions with either the nonpolar tails through dispersion forces or the cationic head groups of the present surfactants through electrostatic interactions. Hence, a selective adsorption of surfactant molecules on some specific lattice planes of t-Se is probably not an energetically favorable process and that is why only long Se NRs of micrometer dimensions are obtained. Conclusions This study concludes that surfactant-assisted aqueous phase synthesis of PbSe and CuSe nanoparticles produces well-defined platelike morphologies. A stronger hydrophobic surfactant produces thin morphologies in both cases. Selective adsorption of surfactant monolayer on {100} crystal planes determines such morphologies and is clearly evident as a shell around each nanoparticle. PbSe reaction also produces fine long Se nanorods as a reaction byproduct which is not observed in the reaction of CuSe. The present study provides a simple aqueous phase method for the synthesis of fine platelike PbSe and CuSe nanoparticles under mild reaction conditions which can find applications in the semiconductor industry. Acknowledgment. These studies were partially supported by financial assistance from CSIR [ref no: 01(2102)/07/EMRII] and UGC [ref no: 34-323/2008(SR)] New Delhi. Supporting Information Available: Size distribution histograms, TEM images, and XRD patterns are available free of charge via the Internet at http://pubs.acs.org.

References (1) (a) Lee, H.; Kwon, K.-W.; Shim, M. J. Mater. Chem. 2007, 17, 1284. (b) Aalivisatos, A. P. Science 1996, 271, 933. (c) Klimov, V. I. J. Phys. Chem. B 2000, 104, 6112. (2) (a) Klimov, V. I.; Mikhailovsky, A. A.; Xu, S.; Malko, A.; Hollingsworth, J. A.; Leatherdale, C. A.; Eisler, H.-J.; Bawendi,

1822

(3) (4)

(5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15)

Crystal Growth & Design, Vol. 10, No. 4, 2010

M. G. Science 2000, 290, 314. (b) Wang, C.; Wehrenberg, B. L.; Woo, C. Y.; Guyot-Sionnest, P. J. Phys. Chem. B 2004, 108, 9027. O’Regan, B.; Gr€atzel, M. Nature 1991, 353, 737. (a) Mueller, A. H.; Petruska, M.; Achermann, A. M.; Werder, D. J.; Akhadov, E. A.; Koleske, D. D.; Hoffbauer, M. A.; Klimov, V. I. Nano Lett. 2005, 5, 1039. (b) Dabbousi, B. O.; Bawendi, M. G.; Onitsuka, O.; Rubner, M. F. Appl. Phys. Lett. 1995, 66, 1316. (a) Bruchez, M.; Moronne, M.; Gin, P.; Weiss, S.; Alivisatos, A. P. Science 1998, 281, 2013. (b) Chan, W. C. W.; Nie, S. Science 1998, 281, 2016. Saloniemi, H.; Kanniainen, T.; Ritala, M.; Leskel€a, M.; Lappalainen, R. J. Mater. Chem. 1998, 8, 651. Yu, S.-H.; Yang, J.; Wu, Y.-S.; Han, Z.-H.; Lu, J.; Xie, Y.; Qiana, Y.-T. J. Mater. Chem. 1998, 8, 1949. Zhang, Y.; Qiao, Z.-P.; Chen, X.-M. J. Mater. Chem. 2002, 12, 2747. (a) Korzhuev, A. A. Fiz. Chim. Obrab.-Mater. 1991, 3, 131. (b) Toyoji, H.; Hirohsi, Y. Jpn. Kokai Tokkyo Koho: JP 02 1990, 173, 622. (a) Lakshmikvar, S. T. Sol. Energy Mater. Sol. Cells 1994, 32, 7. (b) Wang, W.; Yan, P.; Liu, F.; Xie, Y.; Geng, Y.; Qian, Y. J. Mater. Chem. 1998, 8, 2321. Ohtani, T.; Motoki, M. Mater. Res. Bull. 1995, 30, 195. Henshaw, G.; Parkin, I. P.; Shaw, G. Chem. Commun. 1996, 1095. Henshaw, G.; Parkin, I. P.; Shaw, G. J. Chem. Soc.: Dalton Trans. 1997, 231. (a) Wang, X.; Li, Y. Chem. Commun. 2007, 2901. (b) Fan, H. Chem. Commun. 2008, 1383. (a) Latham, A. H.; Williams, M. E. Acc. Chem. Res. 2008, 41, 411– 420. (b) Joswig, J. O.; Springborg, M.; Seifert, G. J. Phys. Chem. B 2000, 104, 2617–2622. (c) Zeiri, L.; Patla, I.; Acharya, S.; Golan, Y.; Efrima, S. J. Phys. Chem. C 2007, 111, 11843–11848. (d) Paoli Lacerda, S. H.; Douglas, J. F.; Hudson, S. D.; Roy, M.; Johnson, J. M.; Becker, M. L.; Karim, A. ACS Nano 2007, 1, 337–347. (e) Shevchenko, E. V.; Talapin, D. V.; Murray, C. B.; O'Brien, S. J. Am. Chem. Soc. 2006, 128, 3620–3637. (f) El-Sayed, M. A. Acc. Chem. Res. 2001, 34, 257. (g) Jana, N. R. Angew. Chem. 2004, 116,

Bakshi et al.

(16)

(17) (18) (19)

(20) (21) (22) (23) (24)

(25)

1562. (h) Rogach, A. L.; Nagesha, D.; Ostrander, J. W.; Giersig, M.; Kotov, N. A. Chem. Mater. 2000, 12, 2676. (i) Ni, T.; Nagesha, D. K.; Robles, J.; Materer, N. F.; Mussig, S.; Kotov, N. A. J. Am. Chem. Soc. 2002, 124, 3980. (j) Mamedova, N. N.; Kotov, N. A.; Rogach, A. L.; Studer, J. Nano Lett. 2001, 1, 281–286. (a) Talapin, D. V.; Yu, H.; Shevchenko, E. V.; Lobo, A.; Murray, C. B. J. Phys. Chem. C 2007, 111, 14049–14054. (b) Wang, X.; Xi, G.; Liu, Y.; Qian, Y. Cryst. Growth Des. 2008, 8, 1406–1411. (c) Cho, K. S.; Talapin, D. V.; Gaschler, W.; Murray, C. B. J. Am. Chem. Soc. 2005, 127, 7140–7147. (d) Talapin, D. V.; Black, C. T.; Kagan, C. R.; Shevchenko, E. V.; Afzali, A.; Murray, C. B. J. Phys. Chem. C 2007, 111, 13244–13249. (e) Sashchiuk, A.; Amirav, L.; Bashouti, M.; Krueger, M.; Sivan, U.; Lifshitz, E. Nano Lett. 2004, 4, 159–165. Xie, Y.; Zheng, X.; Jiang, X.; Lu, J.; Zhu, L. Inorg. Chem. 2002, 41 (2), 387–392. Bakshi, M. S.; Thakur, P.; Sachar, S.; Kaur, G.; Banipal, T. S.; Possmayer, F.; Petersen, N. O. J. Phys. Chem. C 2007, 111, 18087. (a) Bai, G.; Wang, J.; Yan, H.; Li, Z.; Thomas, R. K. J. Phys. Chem. B. 2001, 105, 3105. (b) Zana, R.; Benrraou, M.; Rueff, R. Langmuir 1991, 7, 1072. (c) Wettig, S. D.; Verrall, R. E. J. Colloid Interface Sci. 2001, 244, 377. Imae, T.; Ikeda, S. J. Phys. Chem. 1986, 90, 5216. Gong, J.-Y.; Yu, S.-H.; Qian, H.-S.; Luo, L.-B.; Liu, X.-M. Chem. Mater. 2006, 18, 2012. Lisiecki, I.; Pileni, M. P. J. Am. Chem. Soc. 1993, 115, 3887. (a) Bakshi, M. S.; Singh, K.; Kaur, G.; Yoshimura, T.; Esumi, K. Colloids Surf., A 2006, 278, 129. (b) Bakshi, M . S.; Sachar, S.; Singh, K.; Shaheen, A. J. Colloid Interface Sci. 2005, 286, 369. (a) Fan, H.; Wang, Z.; Liu, X.; Zheng, W.; Guo, F.; Quian, Y. Solid State Commun. 2005, 135, 319. (b) Zhang, B.; Dai, W.; Ye, X.; Zuo, F.; Xie, Y. Angew. Chem., Int. Ed. 2006, 45, 2571. (c) Lu, J.; Xie, Y.; Xu, F.; Zhu, L. J. Mater. Chem. 2002, 12, 2755. (d) Ding, Y.; Li, Q.; Jia, Y.; Chen, L.; Xing, J.; Qian, Y. J. Cryst. Growth 2002, 241, 489. (e) Tang, K.; Yu, D.; Wang, F.; Wang, Z. Cryst. Growth Des. 2006, 6, 2159. (f) Mondal, K.; Roy, P.; Srivastava, S. K. Cryst. Growth Des. 2008, 8, 1580. Ma, Y.; Qi, L.; Ma, J.; Cheng, H. Adv. Mater. 2004, 16, 1023.