Effect of Electrostatic Interactions on the Photophysical Properties of

---

J. Phys. Chem. C 2008, 112, 2317-2324

2317

Effect of Electrostatic Interactions on the Photophysical Properties of the Composites of CdTe Nanocrystals and Carbazole-Containing Polymers Haizhu Sun,†,‡ Hao Zhang,† Junhu Zhang,† Yang Ning,† Tongjie Yao,† Xin Bao,‡ Chunlei Wang,† Minjie Li,† and Bai Yang*,† State Key Laboratory for Supramolecular Structure and Materials, College of Chemistry, Jilin UniVersity, Changchun 130012, People’s Republic of China, and College of Chemistry, Northeast Normal UniVersity, Changchun 130024 People’s Republic of China ReceiVed: February 7, 2007; In Final Form: NoVember 26, 2007

We demonstrated in this study that the electrostatic interaction between nanocrystals (NCs) and polymers was dominant in the photophysical properties of their composites. This effect became more pronounced under illumination and in the presence of the carbazole moieties. The mechanism in which carbazole moieties influenced the PL properties was systematically investigated by means of steady-state fluorescence, fluorescence lifetimes, and ESR spectra. Our experimental results showed that the electrostatic interactions between CdTe NCs and polymers brought defects on the surface of NCs and hence led to the involvement of surface states in the carrier recombination process. Therefore, the composites were inclined to be oxidized when their net charges were slightly positive or negative. The optimal light emitting ratio of NCs to polymers was at the threshold point that NCs could be just completely transferred from the aqueous phase to the organic phase because the composite exhibited nearly electric neutrality at such a ratio. On the other hand, carbazole moieties had an important influence on the intrinsic recombination process of the CdTe NC’s core states due to efficient electron transfer at the interface of the NCs and polymers.

1. Introduction Semiconductor nanocrystals (NCs) directly synthesized in an aqueous solution using water-soluble thio compounds as ligands have attracted much attention in recent years. As compared to nonaqueous synthesis, aqueous synthesis is carried out in water, which is an inexpensive, less toxic, and biocompatible process.1,2 In particular, these NCs can be easily modified with various functional groups during or after preparation, which makes them multifunctional.3 However, several problems must be overcome to realize the applications of these NCs in light emitting diodes (LEDs) and photovoltaic (PV) devices.4 First, the colloidal stability of NCs depends on the electrostatic repulsion resulting from stabilizers as well as surplus ligands in the solution, which are destroyed during the process of fabricating NC devices.5 Second, these ligands normally contain alkyl chains such as 3-mercaptopropionic acid (MPA) and thioglycolic acid (TGA), which undoubtedly influences the transport of charges in the devices and results in low efficiencies of LEDs and PV devices. Other problems such as poor processibility and photochemical instability are also confronted in fabricating LED and PV devices. These problems have severely limited the application of aqueous NCs in optoelectronic devices. One of the best methods to solve these problems is to incorporate these NCs into a polymer matrix, which leads to several advantages: (1) It can efficiently limit the Fo¨rster resonance energy transfer between NCs because they have been insulated by polymer chains;3d (2) it improves the processibility of NCs brought by the polymers;3b (3) the protection from the polymer matrix * To whom correspondence should be addressed. E-mail: byangchem@ jlu.edu.cn; tel.: +86-431-85168478; fax: +86-431-85193423. † Jilin University. ‡ Northeast Normal University.

improves the long-term stability of NCs;3b and (4) the transport ability of charges will be enhanced if the polymers belong to conduction polymers.3c,d Intrigued by the advantages mentioned previously, various methods of incorporating NCs into polymers have been developed recently. Direct incorporation of NCs into the polymer matrix is one of the easiest and earliest methods, but this method is more suitable to NCs synthesized in organic solvents because most of polymers, especially conjugation polymers, are inclined to dissolve in organic solvents. In addition, NCs tend to aggregate due to a lack of covalent attachment.6 Although the problem of aggregation can be overcome by in situ synthesis of NCs in the polymer matrix, the photoluminescence (PL) is dramatically quenched as a result of the absence of efficient passivation.7 Therefore, taking these two strategies as references, the desirable method should be the incorporation of preformed NCs into polymers simultaneously with a strong interaction between NC guest and NC polymer host. In this context, the negative charges of aqueous NCs, as the result of ligand-capping, have been handily used to interact either with polymerizable surfactants or directly with positively charged polymers via electrostatic interactions, leading to the formation of NCpolymer composites.3a-d These processes overcome the major problems associated with the miscibility of NCs with a polymer matrix. Moreover, this method improves the facility of the incorporation of NCs into various functional polymers, such as carbazole-containing polymers. Although the process of composite formation via electrostatic interactions does not involve any ligand exchange, which quenches the PL of NCs,8 it has been found that the electrostatic interaction between NCs and polymers has an important influence on the PL properties of the resultant composites. This effect becomes more pronounced with increasing carbazole

10.1021/jp071076l CCC: $40.75 © 2008 American Chemical Society Published on Web 01/29/2008

2318 J. Phys. Chem. C, Vol. 112, No. 7, 2008 moieties under illumination.3c,d However, few studies have reported the relationship between electrostatic interaction and PL properties of NC-polymer composites. As a requirement of application of these composite materials in optoelectronic devices,9 it is important to reveal the photophysical properties of the composites under an electrostatic field and especially how electrostatic interactions influence the PL properties. In this paper, the effect of electrostatic interactions on the PL properties of NC-polymer composites was systematically investigated through varying the ratios of NCs to polymers in the composites. Our experimental results indicated that the electrostatic interaction between CdTe NCs and polymers brought the carrier recombination process on the NC surface, while carbazole moieties had an important influence on the intrinsic recombination process of NC core states. These effects became more obvious when the content of NCs in the composites was 5.4 wt %. The composites possessed good PL properties when the content of the NCs ranged from 8.35 to 12.8 wt %. 2. Experimental Procedures 2.1. Synthesis. 9-Vinylcarbazole (NVK) was recrystallized from methanol twice, and tellurium powder (-200 mesh, 99.8%) was purchased from Aldrich Chemical Corporation. N,NDimethyl-octadecylamine, 4-vinylbenzyl chloride, and 3-mercaptopropionic acid (MPA) were purchased from Acros Corporation. CdCl2 (99+ %) and NaBH4 (99%) are commercially available products. All of the solvents were analytical grade and used as received. Azodiisobutyronitrile (AIBN) was purified by recrystallization from ethanol prior to use. 2.1.1. Preparation of CdTe NCs. The method for the preparation of NaHTe has been described elsewhere.10 Aqueous colloidal CdTe solutions were prepared by adding freshly prepared NaHTe solution to 1.25 × 10-3 mol/L N2-saturated CdCl2 solutions at pH 9.0 in the presence of MPA as the stabilizing agent. The molar ratio of Cd2+/MPA/HTe- was 1:2.4: 0.2. The resulting mixture was then subjected to a reflux that controlled the growth of CdTe NCs. 2.1.2. Synthesis of Carbazole-Containing Polymer.3c The polymerizable surfactant of octadecyl-4-vinylbenzyl-dimethylammonium chloride (OVDAC) was prepared by modifying the method of Hassanein and Ford.11 The synthesis of the carbazolecontaining polymer poly(9-vinylcarbzole-co-octadecyl-4-vinylbenzyl-dimethyl-ammonium chloride) (CPVKOVDAC) has been reported in ref 3c. Briefly, monomers taken in the desired molar ratio were dissolved in toluene with AIBN added as an initiator. The reaction mixture was flushed with nitrogen for 20 min and then heated in an oil bath at 75 °C to initiate polymerization. The reaction was terminated after it was kept at 75 °C under nitrogen for 72 h, and the polymer was precipitated into a large amount of methanol. The product was redissolved in chloroform and reprecipitated in methanol several times. Then, the product was kept at 40 °C under vacuum overnight. FT-IR (KBr): 2923, 2851 (ν, C-H), 1624, 1595 (two aromatic vibrations in carbazole moieties), 1482, 1449 (vibrations in a five-membered ring of carbazole), 1325, 1227 (ν, C-N) cm-1. 1H NMR (CDCl3, TMS), δH/ppm ) 0.882 (t, 3H); 1.019-1.400 (m, 32H, octadecyl); 3.485 (s, 6H, N-CH3); 6.000-8.010 (m, 12H, aromatic H). 2.1.3. Preparation of CdTe NC-Polymer Composites. The polymer was dissolved in dichloromethane solution at a concentration of 10 mg/mL. The polymer solution was added to the CdTe aqueous solution with vigorous stirring. The organic phase was then separated to yield the CdTe NC-polymer composite solution. The dichloromethane solvent was removed

Sun et al. SCHEME 1: (a) Structure of CPVKOVDAC and (b) Structure of POVDAC

under reduced pressure, and the resulting composite solids were kept in vacuum overnight. The composite solid could be dissolved again in dichloromethane for studying the properties of the composite and characterizations. 2.2. Characterizations. FT-IR spectra were recorded using a KBr window on a Nicolet Avatar 360 FT-IR spectrophotometer. 1H NMR spectra of polymers were recorded on a Bruker Avance 500 MHz NMR spectrometer. UV-vis spectra were acquired using a Shimadzu 3100 UV-vis- near-IR spectrophotometer. Fluorescence experiments were performed on a Shimadzu RF-5301 PC spectrofluorimeter. A NETZSCH STA 449C thermogravimetric analyzer with a heating rate of 10 °C/ min up to 800 °C was used for the thermal degradation of the polymers under nitrogen. Transmission electron microscopy (TEM) was performed on a JEOL-2010 electron microscope operating at 200 kV. Cyclic voltammetry was carried out on BAS 100W Bioanalytical Systems, Inc. instrument, and the potentials were measured in a standard three-electrode twocompartment cell with a platinum counter electrode and an Ag|Ag+ electrode as the reference electrode. The fluorescence lifetimes of CdTe NCs in aqueous solution and in the composites were measured with a Lecroy Wave Runner 6100 digital oscilloscope (1 GHz) using a 325 nm laser (pulse width of 4 ns) from a Sunlite OPO (Continuum) pumped by a Nd:YAG PP-8000 (355 nm) laser as the excitation source. The quantum yields (QYs) of the CdTe NC-polymer composites were determined using quinine sulfate (10-5 M in 0.5 mol/L H2SO4) as the reference. The absorbance at the excitation wavelength was below 0.1 to avoid any significant reabsorption.12 3. Results and Discussions 3.1. Incorporation of CdTe NCs into Polymer. CdTe NCs were transferred from the aqueous phase to the organic phase by using the polymer CPVKOVDAC (see Scheme 1a and Supporting Information Scheme S1 for the structure of CPVKOVDAC and the process of transfer). The electrostatic interaction between negatively charged CdTe NCs and positively charged moieties in polymers was the driving force for the formation of composites. With CPVKOVDAC capping, the surface of the CdTe NCs was varied from hydrophilic to hydrophobic, resulting in the transfer of NCs from water to the organic phase. Experimentally, 10, 20, 40, 60, 80, and 100 mL of a CdTe NC aqueous solution were added to 10 mL of polymer solution in dichloromethane, respectively, to obtain composites with different CdTe NC contents. The abbreviations of composites, volumes of polymer solution and NC solution used in the preparation of composites, QYs of NCs-polymer composites, and the NC content in the composites are summarized in Table 1. Moreover, it can be observed from TEM images (Figure S1) that there was no obvious aggregation of NCs in the composites, although the content of CdTe NCs reached 28.6 wt %. This means that the electrostatic interaction between

CdTe Nanocrystals and Carbazole-Containing Polymers

J. Phys. Chem. C, Vol. 112, No. 7, 2008 2319

TABLE 1: Component, QYs, and Content of CdTe in Composites A-F

a

abbreviations of composites

A

B

C

D

E

F

volume of polymer solution (mL)a volume of CdTe solution (mL) QYs of CdTe NCs in the compositesb content of CdTe in the composites (wt %)c

10 10 0.0096 5.40

10 20 0.045 5.80

10 40 0.16 8.35

10 60 0.11 12.8

10 80 0.09 14.4

10 100 0.052 28.6

Polymer is CPVKOVDAC, and concentration is 10 mg/mL. b Using quinine sulfate (10-5 M) as PL reference. c Determined by TGA.

Figure 1. PL intensities vs the content of CdTe NCs in the composites (excited at 400 nm). The PL spectra were measured under the condition that all samples possessed the same absorbance of NCs, which made it possible to compare the properties of CdTe NCs in the composites with the same concentration. Insets of Figure 1 are the photos of composites A, D, and E whose NCs were transferred from the aqueous phase to the dichloromethane phase.

negatively charged NCs and positively charged polymers could effectively stabilize NCs in polymers. 3.2. Photophysical Properties of CdTe NC-Polymer Composites. To avoid concentration quenching, the powder of composites A-F was dissolved in dichloromethane at a concentration of 0.1 mg/mL. As shown in Figure 1, the PL intensity of the composites had a maximum when the CdTe content was in the range of 8.35-12.8 wt %, from C to D in Figure 1. The inset of Figure 1 shows the optical photos of the solutions in which CdTe NCs were transferred from the aqueous phase to the dichloromethane phase. When the volume of the NC aqueous solution was higher than 60 mL (the CdTe content in the composites was 12.8 wt %), NCs could not be completely transferred to the organic phase. We called the NC content in the range of 8.35-12.8 wt % the threshold value because NCpolymer composites had the strongest PL in this range, which corresponded to the point that CdTe NCs were just completely transferred from the aqueous phase to the organic phase. Although NCs could not be completely transferred to the organic phase for composites E and F, it did not mean that the solution reached its solubility limit because the content of CdTe NCs still increased to 14.4 wt % for composite E and 28.6 wt % for composite F. This result implied that the interaction between NCs and polymers might be different for composites with different NC contents, which should in turn affect their optoelectronic properties.13 It can be observed from Figure 2 that the 1s-1s electron transitions of CdTe NCs for composites A and B were less obvious than those of composites C and D in the UV-vis spectra. The PL spectra of composites A and B also exhibited a wider full width at half-maximum (fwhm) than composites C

and D. The low QYs (Table 1) of A and B further demonstrated that the PL of CdTe NCs had been severely quenched at such ratios. For composites A and B (from Table 1, the contents of CdTe NCs in composites A and B were 5.40 and 5.80 wt %), the relative content of the polymer was higher than that of other composites, which made composites slightly possess positive net charges. This enhanced the polarity of the composites and made them easier to be photooxidated.14 The higher polymer content also led to higher carbazole moieties in the composites, which would in turn shorten the distance between NCs and carbaozle moieties. Therefore, a more efficient electron transfer occurred at the interface of them because the efficiency of electron transfer was an inverse ratio to the distance of exponent power. These two factors were responsible for the decrease of the PL in the composites. The electron-transfer process is discussed in detail in section 3.8. Although the 1s-1s electron transitions of composites E and F became more obvious than those of composites A and B, they were still not as good as those of composites C and D. This is mainly because parts of the composites slightly possessed negative net charges with increasing NC content, which made the composites more easily donate electrons to the accepting electron groups such as oxygen. Therefore, the composites could also be oxidized under negative electrostatic fields. As compared to other composites, composites C and D indicated a pronounced 1s-1s electronic transition in the UV-vis absorption spectra. The PL spectra of composites C and D also exhibited a much narrower fwhm and more symmetry peaks than other composites. Moreover, the QYs (Table 1) of C and D were the highest in the composites. These results indicate that the ratios of NCs to polymers were proper for composites C and D. At such ratios, the capping of polymers provided a protecting shell around the NCs, and the composite showed a nearly electric neutrality in which the photophysical properties were little affected. 3.3. Fluorescence Lifetime of CdTe NC-Polymer Composites. The decay curve and fluorescence lifetimes of composites A-F are indicated in Figure 3. The black circles are experimental data, and they were very well-fitted by the biexponential function A1 exp(t/τ1) + A2 exp(t/τ2) (solid red line). The shorter lifetime (τ1) was attributed to the intrinsic recombination of core states of CdTe NCs, while the longer lifetime (τ2) was attributed to the involvement of surface states in the carrier recombination process.16 Figure 3 clearly indicates that both the intrinsic recombination of core states and the recombination process of surface states showed changes from composites A-F. These results are consistent with the aforementioned PL intensity and QYs (Figure 1 and Table 1). 3.4. Electrochemical Behavior of CdTe NC-Polymer Composites. Figure 4 shows the cyclic voltammograms of composites A-F. Two anodic peaks primarily observed in each curve belong to CdTe NCs in the composites. It has been systematically studied that the peak (asterisked in Figure 4) was related to the oxidation of surface defects by forming intraband gap surface states. The surface defects could be associated with both Cd and Te dangling bonds on the NC surface where the oxidation of CdTe NCs began.15 Therefore, we can judge as to

2320 J. Phys. Chem. C, Vol. 112, No. 7, 2008

Sun et al.

Figure 2. UV-vis absorption and PL spectra (excited at 400 nm) of composites A-F.

whether the composites were easily oxidized through the shift of this peak. As compared to the cyclic voltammogram of aqueous CdTe NCs where the peak was from 0.3 to 0.4 V, the positions of peaks of CdTe NCs in the composites were shifted to the positive direction. This result revealed the fact that coating long alkyl chains of polymers on the surface of CdTe NCs led to the formation of polymer shells, which protected the NCs from photooxidation. Moreover, it could be observed that the composites with more positive net charges were the easiest to be photooxidated, whereas the composites with electric neutrality were the most stable. 3.5. Stability of CdTe NC-Polymer Composites. To study the stability of CdTe NCs in the composites, the PL spectra of composites A, C, D, and F were measured at regular intervals. All of the composites were stored under air and light. Figure 5 shows that the PL intensity for all of the composites decreased with increasing storage time. The slope of the line in Figure 5 represents the decreasing rate of PL intensity. The decreasing rate of PL intensity of composites A and F was much faster than that of composites C and D. All of the cases coincided with the results we obtained previously. In addition, the photophysical behaviors of the CdTe NC-polymer composites were complex as they are related to charges, oxygen, and illumination as well as to carbazole moieties. 3.6. Electrostatic Interaction between CdTe NCs and Polymers. Composites G-L of CdTe NC-poly(octadecyl-4vinylbenzyl-dimethyl-ammonium chloride) (POVDAC) (Scheme 1b) were prepared to study the electrostatic interaction between CdTe NCs and polymers (Table 2). The ratios of CdTe NCs to polymers were fixed in composites G-L. The structure of POVDAC was the same as that of CPVKOVDAC except that CPVKOVDAC contained carbazole moieties. The positive charges in POVDAC were calculated to remain equal to those in CPVKOVDAC to obtain a comparison with CPVKOVDAC. The fluorescence lifetimes of composites G-L are summarized

in Table 2. They were fluorescence lifetimes of CdTe NCPOVDAC stored under oxygen, dark and oxygen, air and light, dark, nitrogen, and dark and nitrogen, for a week, respectively (refer to Supporting Information Figure S2 for the corresponding decay curves). It can be observed from Table 2 that the intrinsic fluorescence lifetimes of the composites (10 to ∼11 ns) were much shorter than those of aqueous CdTe NCs (50 ns) of which the fluorescence lifetime was a single-exponential decay process from intrinsic recombination of the core states. The change of the recombination process from single-exponential decay to biexponential decay clearly indicated that the formation of the composites of NCs and POVDAC introduced defects on the surface of NCs and hence the carrier recombination process. Therefore, the formation of composites could be separated into two steps. In the first step, CdTe NCs combined with POVDAC at the interface of the two phases via electrostatic interactions, and the negative charges on the surface of CdTe NCs were neutralized by the positive charges of POVDAC, leading to a decrease of the fluorescence lifetime of the core states. This step also brought defects on the NC surfaces. As a result, the surface states are involved in the carrier recombination process, represented by the observation of fluorescence lifetime from the surface state.14a,16 In the second step, the long alkyl chains were coated on the NC surface, protecting the NCs from photooxidation. Although the composites were stored under different storage conditions after the formation of composites, they had similar fluorescence lifetimes for both intrinsic recombination of core states and recombination process of surface states. 3.7. Effect of Oxygen Moieties. The fluorescence lifetimes of the composites M-R are also summarized in Table 2. They are fluorescence lifetimes of CdTe NC-CPVKOVDAC stored under oxygen, dark and oxygen, air and light, dark, nitrogen, and dark and nitrogen for a week, respectively (refer to Supporting Information Figure S3 for the corresponding decay

CdTe Nanocrystals and Carbazole-Containing Polymers

J. Phys. Chem. C, Vol. 112, No. 7, 2008 2321

Figure 3. Lifetime decays of composites A-F. Black circles: experimental data. Red line: fitted by biexponential function.

Figure 4. Voltammograms of composites A-F in dichloromethane. Inset: an enlargement of the voltammograms.

curves). The ratios of CdTe NCs to polymers were fixed in the composites M-R. As reported before,2c oxygen had two effects on the PL properties of CdTe NCs. One was that oxygen enabled the formation of CdO on the surface of CdTe NCs, which made the PL intensity of the CdTe NCs increase; the other was that oxygen enabled the CdTe NCs to be oxidized with the cooperation of light. These two effects made the fluorescence

lifetime a little shorter than that of CdTe NCs in aqueous solutions. In our studies, the fluorescence lifetimes of composites M and N confirmed the results that oxygen had little effect on the PL properties of the composites. As shown in Figure 6, we operated composite A under air and light, dark, and dark and nitrogen, respectively. It can be observed that the decreasing rate of PL intensity became much slower for composite A under dark and dark and nitrogen than under air and light. The

2322 J. Phys. Chem. C, Vol. 112, No. 7, 2008

Sun et al.

Figure 5. PL intensities vs storage time of composites A, C, D, and F under air and light (excited at 400 nm).

TABLE 2: Component, Stored Conditions, and Fluorescence Lifetimes of Composites G-R fluorescence lifetime (ns)a composite

stored conditions

polymers usedb,c

τ1

τ2

G H I J K L M N O P Q R

oxygen dark and oxygen air and light dark nitrogen dark and nitrogen oxygen dark and oxygen air and light dark nitrogen dark and nitrogen

POVDAC POVDAC POVDAC POVDAC POVDAC POVDAC CPVKOVDAC CPVKOVDAC CPVKOVDAC CPVKOVDAC CPVKOVDAC CPVKOVDAC

11 10 11 11 11 11 8 9 5 9 5 13

27 25 27 28 29 30 16 23 20 20 22 45

a Fitted by biexponential function A1 exp(t/τ1) + A2 exp(t/τ2). The shorter lifetime (τ1) was attributed to the intrinsic recombination of core states of CdTe NCs, while the longer lifetime (τ2) was attributed to the involvement of surface states in the carrier recombination process. b Concentrations of POVDAC and CPVKOVDAC are 3.3 and 10 mg/ mL, respectively, and the volume of the polymer solution is 10 mL. c Volume of CdTe NC aqueous solution is 40 mL for composites G-R.

Figure 6. PL intensities vs storage time of composite A under air and light, dark, and dark and nitrogen (excited at 400 nm).

decreasing rate of PL intensity for composite A under dark was almost the same as the composites under both dark and nitrogen. These results on one hand confirmed the conclusion that oxygen had little effect on the PL properties of the composites, and therefore, we could exclude the effect of oxygen on the PL properties of the composites; on the other hand, the results revealed that the decrease of PL intensity resulted from the

Figure 7. (a) PL spectra of composite dichloromethane solution excited at 325 nm: different concentrations of CPVKOVDAC with the same concentration of CdTe NCs. (b) PL spectra of the different concentrations of CPVKOVDAC excited at 325 nm, indicating that the PL intensity of the polymers increased with increasing polymer concentration without CdTe NCs.

interaction between CdTe NCs and carbazole moieties cooperated by illumination. 3.8. Effect of Carbazole Moieties and Illumination. From the fluorescence lifetimes of composites O-R, it can be observed that τ2 had no significant change in comparison to those of composites I-L, while τ1 exhibited an obvious variation for composites O (from 11 to 5 ns) and Q (from 11 to 5 ns) in the presence of both carbazole and light. For composites P and R, however, there was no obvious change in τ1 due to the absence of light. So, it can be concluded from this result that only with the cooperation of carbazole moieties and light can the intrinsic recombination process of CdTe NC core states be influenced. The same speculation can be drawn from Figure 7. When the concentration of CdTe NCs in the composites remained constant, the PL intensity for both polymers and CdTe NCs decreased with an increasing polymer concentration by the excitation at 325 nm (Figure 7a). Since the concentration of polymer was kept very low (ranging from 1.5 to 2.3 mg/mL), the increase of the polymer concentration could not lead to PL quenching. Figure 7b also indicates that the PL intensity increased with increasing polymer concentration in the absence of CdTe NCs. Consequently, it could be concluded that the PL quenching resulted from the significant electron transfer of a large number electrons instead of radiative recombination. Several groups have reported using amines to monitor the changes in the optoelectronic properties of NCs. Dannhauser et al. observed the emission enhancement of CdS NCs in the presence of a very low concentration of triethylamine, while the PL was quenched in a high concentration of triethylamine.17 The authors suggested that this enhancement was due to the binding of amine to surface defects that otherwise would trap excited electrons and quench fluorescence. El-Sayed et al., however, reported that the addition of butylamine to a solution of colloidal CdSe NCs caused a decrease in fluorescence intensity. They demonstrated n-butylamine occupied hole sites, which blocked the recombination process and therefore the decrease of the density of luminescent centers.18 This yielded the result that the overall fluorescence intensity had decreased because fewer NCs were emitting, but the observed lifetime did not change because the recombination of electrons and holes still underwent their original dynamics. These two cases combined with our results confirmed that the concentration of the donating electron moieties such as thiethylamine, butylamine, and carbazole did have an influence on the PL of NCs.

CdTe Nanocrystals and Carbazole-Containing Polymers

Figure 8. ESR spectra of composite A (black line) and polymer (red line).

However, the carbazole moieties were too large to bind on the surface of the Cd ion in our system, and the fluorescence lifetime changed with different polymer contents in the composites (Figure 3). Therefore, the mechanism by which carbazole moieties affected the PL of CdTe NCs must be different from those of triethylamine and butylamine. Upon irradiation at 325 nm in our system, electrons were excited into the lowest unoccupied molecular orbital (LUMO) of the polymer, and holes were left in the highest occupied molecular orbital (HOMO) of the polymer. The excited electrons returned from the LUMO to HOMO through a radiative process, resulting in a peak at 365 nm in the PL spectra shown in Figure 7. However, the excited carbazole moieties could form carbazole anions and carbazole cations via resonance with carbaozle moieties at the ground state,19 which was confirmed by the weak ESR signals of the polymer under 250 nm light for 30 min (Figure 8). The carbazole anions migrated to the positively charged surface of the CdTe NC-polymer composite first (as mentioned in section 3.2., the parts of composites could have slightly positive net charges) and then shortened the distance between carbazole moieties and CdTe NCs. Carbazole anions donated their electrons to CdTe NCs, and then the donated electrons recombined with the holes in CdTe NCs. The energy was released mainly in nonradiative form because no obvious new peak was observed from Figure 7a. This reduced the probability of recombination of electrons and holes within CdTe NCs. The results of electron transfer were that holes were confined to polymers, while electrons were limited within CdTe NCs, leading to PL quenching of CdTe NCs and a decrease of τ1. The process of electron transfer was also confirmed by the pronounced ESR signals of composites under 250 nm light for 30 min (Figure 8). The more polymer there was, the more positive net charges and carbazole moieties there were. Therefore, the electron transfer became more efficient with the decrease of NC content. According to the aforementioned analysis, white light emission might be obtained through using the electron transfer from carbazole to CdTe NCs with a proper ratio of NCs to polymers (see Supporting Information Figure S4). 4. Conclusion In summary, we proved that the electrostatic interaction between CdTe NCs and polymers could dramatically influence the photophysical properties of the resultant NC-polymer composites. This influence was mainly from the NC surface

J. Phys. Chem. C, Vol. 112, No. 7, 2008 2323 state that was generated by the electrostatic interaction between NCs and polymers and, therefore, the surface state-related carrier recombination process. This combination made CdTe NCs easily oxidized when the composites possessed positive or negative net charges. In this context, the cooperation of carbazole moieties and light was required to change the intrinsic recombination of core states of CdTe NCs through electron transfer between NCs and carbaozle moieties. Experimental results indicated that the PL intensity of the composites had a maximum when the CdTe content was in the range of 8.35-12.8 wt %, whereas the process of electron transfer was the most efficient when the NC content was below 5.4 wt%. This clearly indicates the possibility of preparing a series of NC-polymer composites with distinct properties and applications just by tuning the ratios of NCs to polymers, for instance, LED applications with proper NC contents and photovoltaic applications with low NC contents. Acknowledgment. This work was financially supported by the Special Funds for Major State Basic Research Projects (2002CB613401 and 2007CB936402), the NSFC-RGC project 20731160002, the National Natural Science Foundation of China (Grant 20534040), the Program for Changjiang Scholars and Innovative Research Team in University (IRT0422), the Program of Introducing Talents of Discipline to Universities (B06009), and the Science Foundation for Young Teachers of Northeast Normal University (20070306 and 20050304). The authors thank Prof. Zhi Yuan Wang for his valuable discussions and editing suggestions. Supporting Information Available: Schematic illustration of transfer process of CdTe NCs from aqueous phase to organic phase, TEM images of CdTe NCs in the composites, lifetime decay of composites G-R, and EL spectrum and I-V curve of the device. This information is available free of charge via the Internet at http://pubs.acs.org. References and Notes (1) (a) Bruchez, M., Jr.; Moronne, M.; Gin, P.; Weiss, S.; Alivisatos, A. P. Science (Washington, DC, U.S.) 1998, 281, 2013-2016. (b) Chan, W. C. W.; Nie, S. Science (Washington, DC, U.S.) 1998, 281, 2016-2018. (c) Mattoussi, H.; Mauro, J. M.; Goldman, E. R.; Anderson, G. P.; Sundar, V. C.; Mikulec, F. V.; Bawendi, M. G. J. Am. Chem. Soc. 2000, 122, 12142-12150. (d) Mamedova, N. N.; Kotov, N. A.; Rogach, A. L.; Studer, J. Nano Lett. 2001, 1, 281-286. (e) Wang, D.; Rogach, A. L.; Caruso, F. Nano Lett. 2002, 2, 857-861. (f) Schroedter, A.; Weller, H.; Eritja, R.; Ford, W. E.; Wessels, J. M. Nano. Lett. 2002, 2, 1363-1367. (2) (a) Zhang, H.; Wang, D.; Yang, B.; Mo¨hwald, H. J. Am. Chem. Soc. 2006, 128, 10171-10180. (b) Kapitonov, A. M.; Stupak, A. P.; Gaponenko, S. V.; Petrov, E. P.; Rogach, A. L.; Eychmu¨ller, A. J. Phys. Chem. B 1999, 103, 10109-10113. (c) Liu, Y.; Chen, W.; Joly, A. G.; Wang, Y.; Pope, C.; Zhang, Y.; Bovin, J.; Sherwood, P. J. Phys. Chem. B 2006, 110, 16992-17000. (3) (a) Zhang, H.; Cui, Z.; Wang, Y.; Zhang, K.; Ji, X.; Lu¨, C.; Yang, B.; Gao, M. AdV. Mater. 2003, 15, 777-780. (b) Zhang, H.; Wang, C.; Li, M.; Zhang, J.; Lu, G.; Yang, B. AdV. Mater. 2005, 17, 853-857. (c) Sun, H.; Zhang, J.; Zhang, H.; Li, W.; Wang, C.; Li, M.; Tian, Y.; Zhang, D.; Chen, H.; Yang, B. Macromol. Mater. Eng. 2006, 291, 929-936. (d) Sun, H.; Zhang, J.; Zhang, H.; Xuan, Y.; Wang, C.; Li, M.; Tian, Y.; Ning, Y.; Ma, D.; Yang, B.; Wang, Z. Y. ChemPhysChem 2006, 7, 2492-2496. (4) (a) Colvin, V. L.; Schlamp, M. C.; Alivisatos, A. P. Nature (London, U.K.) 1994, 370, 354-357. (b) Gao, M.; Lesser, C.; Kirstein, S.; Mo¨hwald, H. J. Appl. Phys. 2000, 87, 2297-2302. (c) Zhao, J.; Bardecker, J. A.; Munro, A. M.; Liu, M. S.; Niu, Y.; Ding, I.; Luo, J.; Chen, B.; Jen, A. K.-Y.; Ginger, D. S. Nano Lett. 2006, 6, 463-467. (d) Steckel, J. S.; Snee, P.; Coe-Sullivan, S.; Zimmer, J. P.; Halpert, J. E.; Anikeeva, P.; Kim, L.; Bulovic, V.; Bawendi, M. G. Angew. Chem., Int. Ed. 2006, 45, 57965799. (e) Robel, I.; Subramanian, V.; Kuno, M.; Kamat, P. V. J. Am. Chem. Soc. 2006, 128, 2385-2393. (f) Snaith, H. J.; Whiting, G. L.; Sun, B.; Greenham, N. C.; Huck, W. T. S.; Friend, R. H. Nano Lett. 2005, 5, 16531657. (5) Tang, Z.; Ozturk, B.; Wang, Y.; Kotov, N. A. J. Phys. Chem. B 2004, 108, 6927-6931.

2324 J. Phys. Chem. C, Vol. 112, No. 7, 2008 (6) Dabbousi, B. O.; Bawendi, M. G. Appl. Phys. Lett. 1995, 66, 13161318. (7) Wang, S.; Yang, S.; Yang, C.; Li, Z.; Wang, J.; Ge, W. J. Phys. Chem. B 2000, 104, 11853-11858. (8) (a) Lee, J.; Sundar, V. C.; Heine, J. R.; Bawendi, M. G.; Jensen, K. F. AdV. Mater. 2000, 12, 1102-1105. (b) Gaponik, N.; Talapin, D. V.; Rogach, A. L.; Eychmu¨ller, A.; Weller, H. Nano Lett. 2002, 2, 803-806. (c) Sill, K.; Emrick, T. Chem. Mater. 2004, 16, 1240-1243. (9) (a) Franzl, T.; Klar, T. A.; Schietinger, S.; Rogach, A. L.; Feldmann, J. Nano Lett. 2004, 4, 1599-1603. (b) Mamedov, A. A.; Belov, A.; Giersig, M.; Mamedova, N. N.; Kotov, N. A. J. Am. Chem. Soc. 2001, 123, 77387739. (c) Franzl, T.; Shavel, A.; Rogach, A. L.; Gaponik, N.; Klar, T. A.; Eychmu¨ller, A.; Feldmann, J. Small 2005, 1, 392-395. (10) Zhang, H.; Zhou, Z.; Yang, B.; Gao, M. J. Phys. Chem. B 2003, 107, 8-13. (11) Hassanein, M.; Ford, W. T. J. Org. Chem. 1989, 54, 3106-3113. (12) Qu, L; Peng, X. J. Am. Chem. Soc. 2002, 124, 2049-2055. (13) Huynh, W. U.; Peng, X.; Alivisatos, A. P. AdV. Mater. 1999, 11, 923-927. (14) (a) Osovsky, R.; Shavel, A.; Gaponik, N.; Amirav, L.; Eychmu¨ller, A.; Weller, H.; Lifshitz, E. J. Phys. Chem. B 2005, 109, 20244-20250. (b)

Sun et al. Yaroslavov, A. A.; Sinani, V. A.; Efimova, A. A.; Yaroslavova, E. G.; Rakhnyanskaya, A. A.; Ermakov, Y. A.; Kotov, N. A. J. Am. Chem. Soc. 2005, 127, 7322-7323. (c) Shevchenko, E. V.; Talapin, D. V.; Murray, C. B.; O’Brien, S. J. Am. Chem. Soc. 2006, 128, 3620-3637. (d) Tian, L.; Zhang, W.; Yang, B.; Lu, P.; Zhang, M.; Lu, D.; Ma, Y.; Shen, J. J. Phys. Chem. B 2005, 109, 6944-6947. (15) Poznyak, S. K.; Osipovich, N. P.; Shavel, A.; Talapin, D. V.; Gao, M.; Eychmu¨ller, A.; Gaponik, N. J. Phys. Chem. B 2005, 109, 1094-1100. (16) Wang, X.; Qu, L.; Zhang, J.; Peng, X.; Xiao, M. Nano Lett. 2003, 3, 1103-1106. (17) Dannhauser, T.; O’Neil, M.; Johansson, K.; Whitten, D.; McLendon, G. J. Phys. Chem. 1986, 90, 6074-6076. (18) (a) Landes, C.; Burda, C.; Braun, M.; El-Sayed, M. A. J. Phys. Chem. B 2001, 105, 2981-2986. (b) Landes, C. F.; Braun, M.; El-Sayed, M. A. J. Phys. Chem. B 2001, 105, 10554-10558. (19) (a) Chen, Y.; Cai, R.; Huang, Z.; Bai, X. Solid State Commun. 1996, 97, 239-242. (b) Wu, W.; Zhang, S.; Li, Y.; Li, J.; Liu, L.; Qin, Y.; Guo, Z. X.; Dai, L.; Ye, C.; Zhu, D. Macromolecules 2003, 36, 62866288. (c) Rathore, R.; Abdelwahed, S. H.; Kiesewetter, M. K.; Reiter, R. C.; Stevenson, C. D. J. Phys. Chem. B 2006, 110, 1536-1540.