10602
J. Phys. Chem. C 2008, 112, 10602–10608
ARTICLES Luminescent Cadmium Sulfide Nanochains Templated on Unfixed Deoxyribonucleic Acid and Their Fractal Alignment by Droplet Dewetting Cunwang Ge,†,‡ Min Xu,† Jinhuai Fang,‡ Jianping Lei,† and Huangxian Ju*,† Key Laboratory of Analytical Chemistry for Life Science (MOE), School of Chemistry and Chemical Engineering, Nanjing UniVersity, Nanjing 210093, P.R. China, School of Chemistry and Chemical Engineering, Nantong UniVersity, Nantong 226007, P.R. China ReceiVed: January 7, 2008; ReVised Manuscript ReceiVed: April 28, 2008
A method for self-assembly of cadmium sulfide (CdS) nanochains (NCs) was proposed by direct deposition of CdS on unfixed DNA (DNA) template with 2-aminoethanethiol as a capping agent. The fractal alignment of CdS NCs on meniscus was then addressed utilizing droplet dewetting. The morphologies of CdS NCs were characterized with transmission electron microscopy, and their luminescent behaviors were investigated by photoluminescence and electrogenerated chemiluminescence (ECL). The presence of DNA scaffolds resulted in perfect surface status and efficient ECL of CdS NCs. The ECL mechanism and the heterogeneous nucleation process were also discussed. Adjusting the molar ratio of DNA to cadmium cation for synthesis could efficiently control the eventual NCs fractal on meniscus during the droplet dewetting and the properties of obtained NCs. The preferred pattern was formed at the molar ratio of 1:1. The present strategy provides a new way to synthesize and align nanowire devices without fixation of DNA and presynthesis of quantum dots (QDs) and has a potential applicability in pattern manufacture in natural way. 1. Introduction The challenge for the realization of functional nanostructure is to find a novel method to hierarchically assemble the nanopatterns. The “bottom-up” assembly using DNA (DNA) template for preparation of nanostructure is currently attracting enthusiastic interest as an alternative to the more popular lithographic1 and masking2 procedures due to the unique recognition, catalytic, and inhibition properties of DNA.3 This strategy has been applied in the generation of artificial DNA superstructures,4 the metallization of DNA,5 and the molecular recognition of DNA.6 The integration of semiconductor quantum dots (QDs), which modulate the electronic and photonic properties across interfaces,7a,b with DNA may yield novel hybrid nanodevices with synergetic functions. Two main approaches for conjugation of semiconductor QDs to DNA on the surface have been achieved. The first method is to assemble presynthesized QDs on the planar DNA-amphiphilic polycation complexes of Langmuir-Blodgett films for forming CdS, CdSe and CdSe/ZnS nanochains (NCs).8a–d The second is the in situ assembly of the NCs, which is performed on the surface of immobilized DNA via the morphology complementarity in the presence of chalcogenide.9a–c Both methods need aforehand preparation of a Langmuir-Blodgett film of DNA or QDs. Herein, as a development of single molecule manipulation technique, such as molecular combing10a and droplet method based on receding meniscus of a drying droplet10b for DNA alignment, a novel strategy was proposed for preparing the linear * To whom correspondence should be addressed. Tel./Fax: +86 25 83593593. E-mail:
[email protected]. † Nanjing University. ‡ Nantong University.
beadlike CdS NCs in aqueous medium by using nonfixed DNA as a soft scaffold. The CdS NCs consisted of a number of small crystal beads in orthorhombic primitive structure. This method simplified greatly the preparation procedure of semiconductor NCs without fixation of DNA and presynthesis of QDs. Although it was reported that the electrogenerated chemiluminescence (ECL) of bare CdS QDs is undetectable because of the instability of intermediate species,11 the ECL emission has been observed from the spherical assemblies of CdS QDs12 and can be enhanced by carbon nanotubes.13 In this work, DNA scaffolds provided the structural origin for the formation of CdS NCs, which caused perfect surface status, and thus led to not only the efficient photoluminescence (PL) but also the strong ECL emission of CdS NCs. In order to employ CdS NCs in constructing optoelectronic devices, the alignment and controlled positioning of the NCs, especially the assembly of NCs between targeted points on DNA-based nanodevices, are highly desirable. A principal step in this procedure is to transfer the NCs to a substrate from their stock solution. This work synthesized NCs in solution system and used a droplet dewetting technique to directly assemble them on a substrate. This technique can predict the deposition of solute and control the shape and density without knowing the chemical nature of the liquid, solute or substrate.14 The commonplace capillary interaction provides a simple and robust route to account in a natural way for the nearly complete transport of the superfluous solute to the periphery. Thus, the fractal alignments of as-prepared CdS NCs were achieved in the radial and the center during drying receding as a result of evaporationinduced capillary flow. The fractal dimensions (FD) analysis with a box-counting model provided the interpretation of the
10.1021/jp8001178 CCC: $40.75 2008 American Chemical Society Published on Web 07/01/2008
Luminescent Cadmium Sulfide Nanochains
J. Phys. Chem. C, Vol. 112, No. 29, 2008 10603
alignment mechanism of the NCs. The present strategy had a potential applicability in pattern manufacture in natural way. 2. Experimental Methods 2.1. Reagents. Calf thymus DNA (poly dispersed solid fibrous form, OD260/OD280 > 1.8) was purchased from Sigma. The molecular mass of DNA, estimated by gel electrophoresis in 1% agarose gel, was 4∼6 × 106 Da. 2-Aminoethanethiol (AET) (Sigma) and tetrabutylammonium perchlorate (TBAP) (Acros) were used without further purification. One millimolar 3-(2-Aminoethylamino)-propyl-methyldimethoxysilane (Fluka) in toluene was used for assembling a hydrophobic glass slices. Other chemicals were of analytical grade. Millipore water (sterilized in Milli-Q plus) with the conductivity of 18 MΩ cm was used in the whole experiment, and all solutions were thoroughly deaerated by bubbling with nitrogen to prevent oxidation. 2.2. Instruments. The morphology and sizes of CdS NCs were characterized by transmission electron microscopy (TEM, JEM-1230, JEOL, operated at 100 kV) and high-resolution TEM (HRTEM, Tecnai G2 F20 S-TWIN, FEI Inc., U.S.A., operated at 200 KV) equipped with energy-dispersive X-ray spectrometer (EDS) (GENESIS 2000 XMS, EDAX Co., U.S.A.). Samples for TEM were prepared by lifting a standard copper grid coated with Formvar film from the solution of CdS NCs, followed by spreading the NCs with filter paper and drying the copper grid at room temperature. The NCs could align in an order direction perpendicular to the drying front due to the capillary force.15 The UV-vis absorption spectra were acquired with a UV-3600 spectrophotometer (Shimadzu Co., Japan). PL spectra were recorded on an LS-55 luminescence spectrophotometer (PerkinElmer, USA). Rhodamine B with a PL quantum yield (QY) of 89% was used as a standard for determining the roomtemperature PL quantum yields of CdS NCs following the previously reported procedure.16 The ECL experiments were performed on a MPI-A multifunctional electrochemiluminescence analytical system (Xi’an Remex Electronic & Technological Co., Xi’an, China) with a three-electrode system comprising platinum wire as counter, Ag/AgCl (3.0 M KCl) for aqueous solution or Ag wire for nonaqueous solution as reference, and a CdS NCs modified carbon paste electrode (CPE) as working electrodes. The ECL emission was detected with a model BPCL ultraweak chemiluminescence analyzer (Institute of Biophysics, Beijing, China) in a pulse mode at a voltage of 800 V, which was sensitive to photons with a wavelength range of 200-800 nm, and 0.1 M TBAP in dichloromethane or 0.1 M pH 7.4 phosphate buffer saline (PBS) containing 0.1 M KNO3 as supporting electrolyte. The zeta potential of CdS NCs was measured on Zetaplus (Brookhaven Instruments Corporation). Each sample was repeatedly measured three times, and the values reported were the mean for two replicate samples. Fractal alignments of CdS NCs were observed using an inverted eclipse-fluorescence microscope (Nikon, Eclipse TE 2000-U, Japan) with a 20× objective. A 100 W mercury lamp was used in combination with a U2MWB excitation cube. The fluorescent images were captured by a cooled CCD camera (DS-U1, Nikon). All the images were stored in a computer and transfer into black and white image with photoshop CS. Fractal dimensions of the structures during assembly were calculated with a fractal analysis system (Ver.3.4.7, donated by National Agriculture and Food Research Organization of Japan). The fractal dimension was calculated using box-counting algorithm with the box values of 2, 4, 8, 16, 32, 64, 128, 256, and 512.17 In this algorithm,
Figure 1. Schematic depiction of the CdS nanochains assembly on nonfixed DNA scaffold.
grids with varying box number were overlaid on the image, and the number of nonempty boxes was counted. Data were plotted as the log of box number of nonempty vs the log of box values. The slope of linear regression corresponded to the fractal dimension (FD). 2.3. Preparation of CdS NCs on Directed DNA. The assembly strategy of CdS NCs was based on direct deposition including four steps in stirring oxygen-free atmosphere, as schematically illustrated in Figure 1. AET and sodium sulfide were used as protective reagent and sulfide source, respectively. The molar ratios of Cd2+:AET:S2- were 1:2.45:0.5, which had been used in the earlier stage of CdS QDs preparation.18 The molar ratio of DNA to Cd2+ was optimized by observing the morphology and PL of the obtained CdS NCs at the DNA concentration of 4.6 × 10-4 mol L-1 (per nucleotide phosphate). Typically, 5.0 mL of 1.84 mM DNA in Tris buffer (10 mM Tris and 0.1 M NaCl) was warmed in a water bath at 37 °C for 30 min to relieve the coil of DNA chain. Under nitrogen atmosphere, 4.6 mL cadmium chloride solution (2.0 mM, 1:1 molar ratio of DNA to Cd2+) was then injected dropwise into the solution with a microsyringe (MDN-0250, Bioanalytical Systems Inc.) to introduce Cd2+ ions to the DNA chains via electrostatic interaction. To avoid the precipitation of the Cd2+ ions conjugated DNA chains, 3.77 mL of 6 mmol L-1 AET (22.5 µmol) as protective reagent was added dropwise with vigorous stirring. After continuous stirring for 1 h, 0.77 mL of freshly prepared oxygen-free sodium sulfide (6 mmol L-1, 4.6 µmol) was introduced into this system. Afterward, CdS NCs were formed gradually under stirring overnight at room temperature. The formed yellowish colloid was dialyzed exhaustively against water to relieve the free cadmium ions and AET. 2.4. Preparation of Modified Electrode for ECL Study. The CdS NCs modified CPE was prepared according to a previous report.19 Briefly, 50 mg graphite powder was thoroughly blended with CdS NCs solution (containing ∼2 mg DNA) followed by evaporation of the water at 4 under vacuum pressure. Subsequently, 18 µL paraffin oil was mixed with above graphite powder. The mixture was finally packed into an electrode cavity of Teflon (3-mm diameter, 5-mm depth). Electrical contact to the paste was established by forcing a copper rod into the the back of the mixture. After being polished manually on a weighing paper, the modified electrode was used as the working electrode in ECL measurements. 3. Results and Discussion 3.1. Influence of Molar Ratio of DNA to Cd2+ on Morphology of CdS NCs. To optimize the molar ratio of DNA to Cd2+, different volumes of 2.0 mM cadmium chloride solution were used to synthesize a series of CdS NCs at the molar ratios of 2:1, 1:1, 3:4, and 1:2. The morphology and size of the as-synthesized CdS NCs were characterized by TEM (details in the Supporting Information). At high molar ratio of DNA to Cd2+ (Figure S1a), the size of formed CdS nanocrystals was ca. 6∼8 nm, and they were discontinuous. With the increasing concentration of Cd2+, the size of CdS nanocrystals increased.
10604 J. Phys. Chem. C, Vol. 112, No. 29, 2008
Figure 2. TEM images of CdS NCs formed at optimized molar ratio of DNA to Cd2+. The inset is HRTEM image of selected area.
At the molar ratio of 1:1, the CdS NCs could be discerned distinctly with linear arrangement on the surface (Figure S1b), the diameter of CdS NCs was about 20∼25 nm, and the beadlike CdS NCs were jointed together to form long NCs, which were similar to a pearly necklace. With the further increase of Cd2+ concentration (Figures S1c and S1d), the formed CdS NCs showed some folding and agglomeration, and the mean diameters for those prepared at two high molar ratios were ca. 25 and 30 nm. Obviously the CdS NCs prepared at the molar ratio of 1:1 showed the best morphology of linear CdS NCs. The structure of CdS NCs was further characterized by HRTEM to provide insight into the structure of linear NCs synthesized at the optimal molar ratio. As shown Figure 2 the TEM and HRTEM images showed chain-like assembly of CdS QDs, and typical products consisted of a large quantity of wirelike structures with the lengths of tens of micrometers, which were different from the monodisperse CdS clusters formed under the same synthesis conditions in the absence of DNA.18 The elemental composition of the CdS NCs was validated to contain N, O, P, S, and Cd elements (shown in the Supporting Information). The appearance of P element should be attributed to DNA, whereas the N element should be attributed to DNA and AET. The TEM results suggested that the morphology of CdS NCs could be controlled by the molar ratio of DNA to Cd2+. This behavior was attributed to the fact that the hydrodynamic force of a receding meniscus could stretch linearly DNA molecules, and the uniform formation of CdS nanobeads on DNA scaffolds accelerated the stretch. The Fourier transform pattern in the selected area of TEM image was concentric rings (not shown), but the lattice fringes of QDs were discerned (inset in Figure 2). It revealed that CdS NCs were consisted of a number of small crystal beads with highly crystalline structure, similar to those obtained with scaffold-directed methods.20 The HRTEM image showed the interplanar distances of 4.75, 3.44, and 2.82 Å in three directions (inset in Figure 2), corresponding to those of 4.77, 3.45, and 2.83 Å indexed in (300), (232), and (314) stacking for CdS in orthorhombic primitive structure (JCPDS 47-1179), which did not present in cubic face-centered or hexagonal primitive-type structure. Here a small deviation from the bulk values for interplanar distances was observed, which originated from the deviation of calculation. The orthorhombic form of CdS stacking was first obtained through solid-gas
Ge et al.
Figure 3. UV-vis (a) and PL emission (b) spectra of CdS NCs at different molar ratios of DNA to Cd2+. The excitation wavelength is 350 nm. Inset in panel a: enlargement of the selected area.
reactions in previous report.21 The structure of presented CdS NCs templated on DNA scaffolds was different from cubic Sphalerite-type (zinc blend) growing in solution,22a,b and hexagonal Wurtzite-type evaporating on vapor-solid,23 respectively. The favored orthorhombic phase of CdS nanocrystals was due to the presence of DNA. The cadmium cations were first pinned to DNA chains through the electrostatic interaction, which led to the formation of orthorhombic cells with lowpacked density. On the other hand, large polarizability of sulfide ion and the radius ratio rCd2+/rS2- of 0.516 (less than 0.732) made the CdS tend to form the structure with low coordination number. 3.2. Spectroscopic Properties of CdS NCs. II-VI and III-V semiconductor quantum dots such as CdS have been known to show the size-tunable atomic-like properties arising from quantum confinement in nanometer scale. Colloidal nanocrystals of these semiconductors display striking absorption and strong fluorescence arising from their “artificial atom” character. Thus, the formation process of the CdS NCs can be identified from UV-vis absorption and PL spectra of the obtained solution. The UV-vis spectra of CdS NCs samples obtained at different molar ratios of DNA to Cd2+ showed maximum absorption at 350 (2:1), 368 (1:1), 372 (3:4), and 378 (1:2) nm, respectively (Figure 3a), which were ascribed to the first excitonic 1Sh-1Se transition.24 At high molar ratio of DNA to Cd2+ (Figure S1a), the size of discontinuous CdS nanocrystals was ca. 6∼8 nm, which was near to the Bohr diameter of exaction of CdS, leading to the greatest blue shift when compared with the individual absorption at 450 and 530 nm for quantum-CdS nanocrystallites stabilized by calf thymus DNA25 and bulk CdS. With the decreasing molar ratio of DNA to Cd2+, the size of CdS nanocrystals increased, and the UV absorption red-shifted, but the maximum adsorption occurred at shorter wavelength than that of bulk CdS due to the reduction of nanoparticle size, which was consistent with well-known quantum confinement in the cluster.26 When compared with bare CdS nanoclusters the red-shifts of about 60 nm were observed.22b The absorption at 260 nm was due to the electronic excitation of nitrogenous bases of DNA. The UV-vis absorption at 260 nm was reinforced with the decrease of the ratio from 2:1 to 1:1. This might originate from the interaction between DNA
Luminescent Cadmium Sulfide Nanochains and Cd2+, which led to a significant distortion of the bases stacking.27 The absorption at 260 nm at the molar ratio of 3:4 decreased to 85% of that at the molar ratio of 1:1, indicating that the aggregation of CdS nanocrystals decreased the absorption. The PL excitation curves of CdS NCs obtained at different molar ratios revealed the broad bands ranging from 310 to 370 nm (in the Supporting Information, Figure S3), while the PL emission bands were observed at ca. 510 nm (Figure 3b), which correlated to electron transition from the excitated states to the surface localized trap states. With the decreasing molar ratio of DNA to Cd2+ the PL spectra showed red-shift from 505 nm (2:1), 509.5 nm (1:1), and 512 nm (3:4) to 515 nm (1:2), which was the characteristic property of quantum confinement effect going with the increase of NCs width. No emission was observable at a wavelength longer than 600 nm, which was different from that of CdS QDs stabilized by DNA with defect emission around 580 nm,25 indicating better surface status of CdS NCs.28 The addition of AET protector prior to the sulfide source was the only difference. It indicated that AET protector played an important role in the aggregation of CdS nanocrystals on scaffold DNA. This might be attributed to the presence of promoted crystallinity with the size regime of quantum confinement, which also could be observed from the HRTEM measurement. The PL emission spectra indicated that the prior addition of AET protector was important for controlling the aggregation of QDs. The spectral measurement showed the strongest PL excitation and emission occurred at the molar ratio of 1:1, suggesting that the proper DNA was responsible for the particular surface photophysics. According to the measurement method reported previously with a standard quantum yield (QY) of 89% for rhodamine B,16 the PL QYs of different CdS NCs were estimated to be 12.4% (2:1), 16.5% (1:1), 11.3% (3:4), and 9.6% (1:2), respectively. The PL emission fwhm of about 117 nm was broader than that of the pure CdS nanocrystals. The aggregation of CdS nanocrystals on DNA scaffolds might be the cause of PL broadening. 3.3. ECL Properties of CdS NCs. The information of surface energy of the CdS NCs can be obtained from their ECL behaviors. Although the CdS NCs modified electrode in a CH2Cl2 solution of 0.1 M TBAP showed a poor cyclic voltammetric behavior in CH2Cl2 system (Figure 4), similar to the case of CdSe NCs,29 two ECL processes could be observed in cathodic scan. The first ECL peak occurred at -0.04 V, and the second ECL emission began at the voltage of -0.32 V, with a plateau occurring at -2.13 V. The intensity of the first emission was much lower than the latter. Meanwhile, the anodic ECL emission started at the voltage of +0.77 V and reached the plateau at the voltage of +1.13 V. The peak-to-peak separation between reduction and oxidation ECL emission was 3.26 V, which was comparable to 3.39 eV calculated from the UV-absorption spectrum at maximum absorption of 368 nm for CdS NCs. Consequently, these cathodic and anodic ECL peaks could be correlated directly to electron transfer at HOMO and LUMO of CdS NCs. As DNA was wrapped inside the CdS NCs,11 DNA could not act as a coreactant for ECL emission. The ECL intensity of the CdS NCs modified electrode in CH2Cl2 was about 7 times larger than that in aqueous solution (see the Supporting Information, Figure S4), similar to the case of bare CdSe nanocrystals,30 in which the solvent CH2Cl2 was considered as a coreactant by the following electrochemical process. At a high potential CH2Cl2 could be oxidized:
J. Phys. Chem. C, Vol. 112, No. 29, 2008 10605
Figure 4. Cyclic scan ECL curve (a) and time-dependent change in ECL intensity for initial three cycles (b) of CdS NCs modified electrode at 400 mV s-1 in CH2Cl2 solution containing 0.1 M TBAP. The arrows indicate the potential scan direction.
TABLE 1: Dependence of Zeta Potential of CdS NCs on Molar Ratio of DNA to Cd2+ sample 1 2 3 4
molar ratio
zeta potential (mV)
half-width (mV)
2:1 1:1 3:4 1:2
0.3 ( 0.3 -11.3 ( 2.9 -34.6 ( 5.2 -29.9 ( 2.0
3.0 3.1 3.9 5.4
CH2Cl2 f CH2Cl2•+ + e-
(1)
The ejected electron was captured by another solvent molecule to form a neutral radical:31
CH2Cl2 + e- f CH2Cl• + Cl-
(2)
Cl•
The neutral radical CH2 could be proposed as the oxidant in this system. When the potential was scanned cathodically, electrons were injected into CdS NCs, and the formed reduced species CdS•- collide with CH2Cl• to produce excited states CdS*, which generated emission as follows:
CdS + e- f CdS••-
•
(3) -
CdS + CH2Cl f CdS + CH2Cl
(4)
CdS* f CdS + hν
(5)
*
The ECL signal of CdS NCs in the cathodic process was very stable. At relatively negative potential, the electron could also be injected into CH2Cl2:
CH2Cl2 + e- f CH2Cl2•-
(6)
Thus, in the anodic process the procreant species CH2Cl2•would collide with the oxidized species CdS•+ to produce excited states CdS* and emission with radical annihilation
CdS•+ + CH2Cl2•+ f CdS* + CH2Cl2
(7)
CdS* f CdS + hν
(8)
The small ECL peak at -0.04 V could be ascribed to the reaction between CdS•+ and CdS•-
10606 J. Phys. Chem. C, Vol. 112, No. 29, 2008
Ge et al.
Figure 5. Fluorescence microscopy topographies of fractal alignment of CdS NCs formed at molar ratios of 2:1 (a), 1:1 (b), 3:4 (c), and 1:2 (d) for DNA to Cd2+ on globular meniscus of drying droplet. The fractal dimensions of CdS NCs were 2.51 (b), 2.43 (c), and 2.44 (d).
CdS•+ + CdS•- f CdS* + CdS
(9)
CdS* f CdS + hν
(10)
Based on no perceptible ECL signal for pure CdS nanocrystals, the efficient and stable ECL could be attributed to the improvement of crystallization and perfect surface status of CdS NCs in presence of DNA scaffolds. 3.4. Mechanism for Formation of the Necklacelike Nanochain. The underlying mechanism for the formation of necklacelike nanochains can be explained with the lateral inhomogeneity of the template surface.32 In principle, Cd2+ cations adsorbed on the DNA surface could serve as nucleation sites for the formation of extremely regular NCs via siteselective nucleation. After AET was injected, both the adsorbed Cd2+ cations on DNA and the free Cd2+ cations coordinated with AET. The overall stability constant of Cd-AET is β ) K1 × K2, where K1 and K2 are equal to 6.92 × 109 and 7.59 × 107, respectively,33 while the solubility product of CdS is 8.0 × 10-27,34 corresponding to the stability constant of 1.25 × 1026, suggesting CdS can be easier formed thermodynamically. Thus, at the initial introduction of sulfide source to this system the Cd-AET adsorbed on the surface of DNA reacted with S2- to form deposited CdS, which interacted further with Cd-AET in solution as “anchor sites”, leading to more Cd-AET to be adsorbed on the surface of DNA. The further substituent of AET by S2- made the coordinating AET molecules be expelled outside for capping the CdS QDs. During the continuously stirring overnight the procreant CdS QDs were accumulated outside along with the development of nucleation. When S2- ions were gradually consumed and the capping AET molecules were released, the interaction between Cd-AET and the accumulated CdS QDs was prevented to form a necklacelike morphology. The AET covered on CdS NCs surfaces led to high stability of the CdS NCs solution.
3.5. Fractal Alignment of CdS NCs by Droplet Dewetting. The as-prepared CdS CNs appeared to easily form clusters, flocks and finally a line pattern during drying receding of a droplet. By controlling the molar ratio of DNA to Cd2+ different patterns of the CdS CNs alignment could be formed on the hydrophobic or hydrophilic glass slide. The inverted eclipsefluorescence microscope was used to study the alignment of CdS NCs on meniscus during drying receding of a droplet. Figure 5 depicts typical images of four types of CdS NCs alignments on the hydrophilic slide. At the molar ratio of 2:1, no obvious CdS NCs were discerned, as confirmed by TEM (Figure S1a), because of the low luminescence (Figure 5a). When the molar ratio decreased to 1:1, the unique pattern of CdS NCs was featured with straight alignment of NCs. However, further decrease of the molar ratio to 3:4 and 1:2, the alignment of CdS NCs showed some divaricable flocks similar with the topology of tree leafs, which was in accordance with the TEM with some folding and agglomeration (Figure S1c and S1d), indicating that the molecules were relaxed in a quasi-2D state. Obviously, the alignment of CdS NCs depended on the molar ratio of DNA to Cd2+ and had linear pattern at the molar ratio of 1:1. Almost all particulates or macroscopic materials in contact with a liquid show electronic charge on their surfaces. Zeta potential is an important and useful indicator of the charge. As shown in Table 1, the zeta potential of the CdS NCs was related to the molar ratio of DNA to Cd2+. At the molar ratios of 1:2 and 3:4 the absolute values of the zeta potentials were much larger than those obtained at 1:1 and 2:1, whereas the absolute value of the zeta potential at 2:1 was much smaller than that obtained at 1:1. The small absolute value of zeta potential led to the agglomeration of the CdS NCs due to the electrostatic repulsion, thus no alignment could be formed at the molar ratio of 2:1. On the other hand, the large absolute value of the zeta
Luminescent Cadmium Sulfide Nanochains
J. Phys. Chem. C, Vol. 112, No. 29, 2008 10607
Figure 6. Fluorescence microscope topographies of CdS NCs aligned on hydrophobic (a) and hydrophilic (b) glass slides in the center of droplet during dewetting.
Figure 7. Fluorescence microscopy topographies of CdS NCs aligned on hydrophilic glass slides from square (a) and trigonal (b) of small droplet during dewetting.
potential indicated great electrostatic repulsion, which inhibited the solvent from carrying the dispersed materials to the perimeter of the droplet and led to aggregation of the solute at the center of meniscus. Thus the appropriate zeta potential was important for forming straight alignment of CdS NCs, which occurred at the molar ratios of 1:1 (Figure 5b). The capillary dewetting may be the genesis of the radial alignment of CdS NCs. In order to elucidate the alignment mechanism of CdS NCs, different hydrophilic surfaces were adopted for depositing the NCs. At the molar ratio of 1:1 the fluorescence microscope images of CdS NCs aligned on hydrophobic and hydrophilic glass slides were showed in Figure 6. On hydrophobic surface the hydrophobic groups, which existed inside the bases of DNA35 to protect Watson-Crick hydrogen bonding of base pairs against water, pinned on the hydrophilic glass slides, and the evaporated liquid of NCs droplet from the center of drying drop could be replenished from the edge. The resulting centripetal flow stretched the CNs from edge to center to form linear structure eventually. On hydrophilic glass slide, after exterior liquid evaporation a contact line of the drying drop was pinned on the substrate, which contained almost all the solute.14 The liquid evaporation from the edge was replenished by liquid from the interior, producing an
outward capillary flow of the solvent and leading to highly selective deposition along the perimeter of the meniscus. Eventually, the characteristic pattern of CdS NCs was aligned in the center of droplet on hydrophilic glass slide. The fractal alignment of CdS NCs could be adjusted dramatically by the capillary direction of flow on hydrophilic surface. The typical fluorescence microscopy topographies of CdS NCs aligned on square and trigonal droplets on hydrophilic glass slides were shown in Figure 7. The fir-leaf-like patterns of CdS NCs were formed in the direction perpendicular to a contact line of the droplet. The alignment of CdS NCs during droplet dewetting demonstrated that the stretching direction of the CNs was always perpendicular to the air/liquid interface. The shape and motion of this interface served as an effective local field directing the CNs alignment dynamically, which was similar to the case in microchannels.36 Since the structure within a pattern of CdS NCs was fractal in nature, fractal dimensions (FDs) were used to quantify the microstructure of CdS NCs alignment. The FDs of NCs alignments shown in Figure 5 could be explained from 2-dimensional image analysis of light micrographs using boxcounting model. Their FDs were 2.51 (b), 2.43 (c), and 2.44 (d), respectively, which were close to the value of 2.5 calculated
10608 J. Phys. Chem. C, Vol. 112, No. 29, 2008 from the diffusion-limited aggregation, suggesting that the alignment of CdS NCs was proceeded through DLA mechanism limited by Brownian diffusion.37 4. Conclusions Using nonfixed DNA as a soft template, beadlike CdS nanochains were successfully synthesized under mild solution conditions. The fractal alignment of CdS NCs was also developed on meniscus with hydrophobic or hydrophilic glass slide utilizing droplet dewetting. The presence of DNA scaffolds resulted in perfect surface status and efficient ECL of CdS NCs. By controlling the molar ratio of DNA to Cd2+, different patterns of CdS NCs and their alignments could be obtained, and the best morphology of linear CdS NCs with short-range ordering nanocrystallite in orthorhombic primitive structure and the strongest PL excitation and emission occurred at the molar ratio of 1:1. The fractal alignment could be adjusted dramatically by the capillary direction of flow and was proceeded through diffusion-limited aggregation mechanism limited by Brownian diffusion. The proposed method for CdS NCs preparation provided a new way to prepare various metal or metal chalcogenides NCs, and the NCs could be potentially applied in both luminescence devices and electronic nanodevices. Acknowledgment. This work was supported by the National Natural Science Foundation of China (Nos. 60571055, 20535010, 20521503, and 90713015), China Postdoctoral Science Foundation (20060400914), and Jiangsu Postdoctoral Science Foundation (0701010). We thank Chunlai Zhu (Jiangsu Key Laboratory of Neurogenerat, Nantong University) and Fengli Bei (Nanjing University of Science & Technology) for TEM and HRTEM measurements, respectively. Supporting Information Available: Optimization of the molar ratio of DNA to Cd2+, energy-dispersive X-ray spectrum and PL excitation spectra of CdS NCs, and cyclic scan ECL curve of CdS NCs in aqueous solution. This material is available free of charge via the Internet at http://pubs.acs.org. References and Notes (1) Hsieh, H. H.; Wu, C. C. Appl. Phys. Lett. 2007, 91, 013502. (2) Becerril, H. A.; Stoltenberg, R. M.; Monson, C. F.; Woolley, A. T. J. Mater. Chem. 2004, 14, 611. (3) Cairns, M. J.; Saravolac, E. G.; Sun, L. Q. Curr. Drug Targets 2002, 3, 269. (4) Fahlman, R. P.; Sen, D. J. Am. Chem. Soc. 1999, 121, 11079. (5) Braun, E.; Eichen, Y.; Sivan, U.; Ben-Yoseph, G. Nature 1998, 391, 775. (6) Williams, K. A.; Veenhuizen, P. T. M.; de la Torre, B. G.; Eritja, R.; Dekker, C. Nature 2002, 420, 761. (7) (a) Gittins, D. I.; Bethell, D.; Schiffrin, D. J.; Nichols, R. J. Nature 2000, 408, 67. (b) Katz, E.; Willner, I. Angew. Chem., Int. Ed. 2004, 43, 6042.
Ge et al. (8) (a) Li, L. S.; Jin, J.; Yu, S.; Zhao, Y.; Zhang, C.; Li, T. J. J. Phys. Chem. B 1998, 102, 5648. (b) Torimoto, T.; Yamashita, M.; Kuwabata, S.; Sakata, T.; Mori, H.; Yoneyama, H. J. Phys. Chem. B 1999, 103, 8799. (c) Artemyev, M.; Kisiel, D.; Abmiotko, S.; Antipina, M. N.; Khomutov, G. B.; Kislov, V. V.; Rakhnyanskaya, A. A. J. Am. Chem. Soc. 2004, 126, 10594. (d) Stsiapura, V.; Sukhanova, A.; Baranov, A.; Artemyev, M.; Kulakovich, O.; Oleinikov, V.; Pluot, M.; Cohen, J. H. M.; Nabiev, I. Nanotechnology 2006, 17, 581. (9) (a) Coffer, J. L.; Bigham, S. R.; Li, X.; Pinizzotto, R. F.; Rho, Y. G.; Pirtle, R. M.; Pirtle, I. L. Appl. Phys. Lett. 1996, 69, 3851. (b) Jin, J.; Jiang, L.; Chen, X.; Yang, W. S.; Li, T. J. Chin. J. Chem. 2003, 21, 208. (c) Dong, L.; Hollis, T.; Connolly, B. A.; Wright, N. G.; Horrocks, B. R.; Houlton, A. AdV. Mater. 2007, 19, 1748. (10) (a) Michalet, X.; Ekong, R.; Fougerousse, F.; Rousseaux, S.; Schurra, C.; Hornigold, N.; van Slegtenhorst, M.; Wolfe, J.; Povey, S.; Beckmann, J. S.; Bensimon, A. Science 1997, 277, 1518. (b) Maubach, G.; Fritzsche, W. Nano Lett. 2004, 4, 607. (11) Bard, A. J.; Ding, Z. F.; Myung, N. Struct. Bonding (Berlin) 2005, 118, 1. (12) Ren, T.; Xu, J. Z.; Tu, Y. F.; Xu, S.; Zhu, J. J. Electrochem. Commun. 2005, 7, 5. (13) Ding, S. N.; Xu, J. J.; Chen, H. Y. Chem. Commun. 2006, 3631. (14) Deegan, R. D.; Bakajin, O.; Dupont, T. F.; Huber, G.; Nagel, S. R.; Witten, T. A. Nature 1997, 389, 827. (15) Bensimon, A.; Simon, A.; Chiffaudel, A.; Croquette, V.; Heslot, F.; Bensimon, D. Science 1994, 265, 2096. (16) Demas, J. N.; Grosby, G. A. J. Phys. Chem. 1971, 75, 991. (17) Smith, T. G.; Lange, G. D.; Marks, W. B. J. Neurosci. Methods 1996, 69, 123. (18) Vossmeyer, T.; Katsikas, L.; Giersig, M.; Popovic, I. G.; Diesner, K.; Chemseddine, A.; Eychmueller, A.; Weller, H. J. Phys. Chem. B 1994, 98, 7665. (19) Ju, H. X.; Liu, S. Q.; Ge, B. X.; Lisdat, F.; Scheller, F. W. Electroanalysis 2002, 14, 141. (20) Xia, Y.; Yang, P.; Sun, Y.; Wu, Y.; Mayers, B.; Gates, B.; Yin, Y.; Kim, F.; Yan, H. AdV. Mater. 2003, 15, 353. (21) Kizilyalli, M.; Bilgin, M.; Usanmaz, A. J. Solid State Chem. 1989, 80, 75. (22) (a) Herron, N.; Calabrese, J. C.; Farneth, W. E.; Wang, Y. Science 1993, 259, 1426. (b) Vossmeyer, T.; Reck, G.; Katsikas, L.; Haupt, E. T. K.; Schulz, B.; Weller, H. Science 1995, 267, 1476. (23) Dong, L.; Gushtyuk, T.; Jiao, J. J. Phys. Chem. B 2004, 108, 1617. (24) Bawendi, M. G.; Wilson, W.; Rothherg, L.; Carroll, P.; Jedju, T.; Steigerwald, M. L.; Brus, L. E. Phys. ReV. Lett. 1990, 65, 1623. (25) Kulkarni, S. K.; Ethiraj, A. S.; Kharrazi, S.; Deobagkar, D. N.; Deobagkar, D. D. Biosens. Bioelectron. 2005, 21, 95. (26) Onoa, G. B.; Cervantes, G.; Moreno, V.; Prieto, M. J. Nucleic Acids Res. 1998, 26, 14730. (27) Steigerwald, M. L.; Brus, L. E. Acc. Chem. Res. 1990, 23, 183. (28) Agata, M.; Kurase, H.; Hayashi, S.; Yamamoto, K. Solid State Commun. 1990, 76, 1061. (29) Myung, N.; Ding, Z.; Bard, A. J. Nano Lett. 2002, 2, 1315. (30) Bae, Y.; Myung, N.; Bard, A. J. Nano Lett. 2004, 4, 1153. (31) Alfassi, Z. B.; Mosseri, S.; Neta, P. J. Phys. Chem. 1989, 93, 1380. (32) Witkowski, N.; Borensztein, Y.; Baudot, G.; Repain, V.; Girard, Y.; Rousset, S. Phys. ReV. B 2004, 70, 085408. (33) Li, Y.; Martell, A. E. Inorg. Chim. Acta 1995, 231, 159. (34) Cao, X. Z.; Zhang, W. H. Inorganic Chemistry, 2nd ed.; High Education Publishing Company: Beijing, 1983; p 625. (35) Pasternack, R. F.; Gibbs, E. J. Biochemistry 1983, 22, 5409. (36) Petit, C. A. P.; Carbeck, J. D. Nano Lett. 2003, 3, 1141. (37) Witten, T. A.; Sander, L. M. Phys. ReV. Lett. 1981, 47, 1400.
JP8001178
