Determination of quantum confinement in CdSe

Determination of quantum confinement in CdSe nanocrystals by cyclic voltammetry Erol Kucur, Jürgen Riegler, Gerald A. Urban, and Thomas Nann Citation: J. Chem. Phys. 119, 2333 (2003); doi: 10.1063/1.1582834 View online: http://dx.doi.org/10.1063/1.1582834 View Table of Contents: http://jcp.aip.org/resource/1/JCPSA6/v119/i4 Published by the AIP Publishing LLC.

Additional information on J. Chem. Phys. Journal Homepage: http://jcp.aip.org/ Journal Information: http://jcp.aip.org/about/about_the_journal Top downloads: http://jcp.aip.org/features/most_downloaded Information for Authors: http://jcp.aip.org/authors

Downloaded 30 Jul 2013 to 37.229.95.32. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://jcp.aip.org/about/rights_and_permissions

JOURNAL OF CHEMICAL PHYSICS

VOLUME 119, NUMBER 4

22 JULY 2003

Determination of quantum confinement in CdSe nanocrystals by cyclic voltammetry Erol Kucur, Ju¨rgen Riegler, Gerald A. Urban, and Thomas Nanna) Freiburg Materials Research Center (FMF), Stefan-Meier-Strasse 21, D-79104 Freiburg, Germany

共Received 28 March 2003; accepted 24 April 2003兲 Ionization potentials I p , electron affinities E.A., and the quantum confinement in CdSe nanocrystals were determined by means of cyclic voltammetry. The results were compared to values obtained from spectroscopic measurements, especially UV/vis absorption and photoluminescence emission spectra. Absolute band gap positions were obtained from the electrochemical measurements and discussed with regard to vacuum level values. The results are in good agreement with theoretical expectations and spectroscopic data. © 2003 American Institute of Physics. 关DOI: 10.1063/1.1582834兴 I. INTRODUCTION

clic voltammetry in combination with UV/vis spectroscopy has been used to estimate the band structure of semiconducting polymer thin films.14,15 Cyclic voltammetry is a dynamic electrochemical method, where current-potential curves are recorded at well-defined scan rates. The measured oxidation potential of an electroactive substance 共e.g., a thin film of conducting polymer兲 correlates directly with the ionization potential I p and the reduction potential with the electron affinity E.A. As long as the vacuum level potentials of the common reference electrodes are known, the band edge positions of electroactive materials can be estimated.16 This kind of information cannot be determined from UV/vis spectroscopy. Compared to methods like x-ray photoelectron spectroscopy or ultraviolet photoelectron spectroscopy, cyclic voltammetry is more easily performed and this electrochemical method is closer to the targeted applications. The present paper describes a similar approach to estimate the band edge positions in semiconductor nanocrystals with different sizes. It will be shown that size dependency of band structures as well as vacuum level values for band edge energies can be determined by cyclic voltammetry of adsorbed nanocrystals.

Essentially, three reasons are responsible for the increasing interest in semiconductor nanocrystals: first, mesoscopic effects can be observed in nanocrystals, which are mostly based upon a confinement of charges into at least one dimension 共a so-called ‘‘quantum size effect’’兲; second, highly monodisperse nanocrystals have been available since the early 1990s; and third, this new class of materials has an extraordinary high potential for applications, particularly within the area of catalysis,1 photovoltaics,2– 4 phosphors,5 light-emitting diodes,6 – 8 and the labeling of biological molecules.9–11 Among the mesoscopic properties of semiconducting nanocrystals, the confinement of charge carriers within these particles is most prominent. Excitons with a size smaller than the Bohr radius can be described with a particle-in-a-box model,12,13 where the particle is represented by the exciton and the box by the nanocrystal. This model and experimental data show that the band gap of semiconductor nanoparticles increases with decreasing crystal size. Therefore, the photoluminescence of direct semiconducting nanocrystals is blueshifted with decreasing particle size. This electro-optical feature is the basis for almost every 共potential兲 application of such particles. Control and knowledge of the band gap and band edge positions are crucial for photovoltaic and electroluminescent applications. The present work was motivated by the objective to develop photovoltaic and electroluminescent devices as well as new building blocks for polymer microelectronics on the basis of nanocrystal/conducting polymer composites. The charge transfer between nanocrystals, conducting polymers, and electrodes is crucial for such devices. In order to enable this charge transfer, the bands of the polymers and nanocrystals have to possess favorable positions. Because these band edge positions are size dependent in case of nanocrystals, a method had to be found to estimate these positions and relate them to the band positions of conducting polymers. Within the research on organic light-emitting diodes, cy-

II. EXPERIMENT A. Chemicals

Spherical CdSe nanocrystals have been prepared similarly to a previously published procedure:17 313.6 mg cadmium naphthenate 共480 ␮mol, 17.2%, Strem chemicals兲 and 4 g tri-n-octylphosphine oxide 共TOPO, Avocado兲 were given in a Schlenk flask within an inert gas atmosphere (N2 ). The mixture was heated to 225 °C and degassed several times. A solution of 31.9 mg selenium 共400 ␮mol, Merck兲 in 2 mL trin-octylphosphine 共Se/TOP兲 was prepared in a drybox and transferred into a syringe. The Se/TOP solution was injected quickly into the cadmium solution under Schlenk conditions, and the mixture was stirred for 5 min at 200 °C. The reaction mixture was cooled down to room temperature and the nanocrystals were precipitated with 10 mL of dry methanol. After centrifugation, the nanoparticles were washed with dry methanol once and redissolved in chloroform for further measurements.

a兲

Author to whom all correspondence should be addressed. Electronic mail: [email protected]

0021-9606/2003/119(4)/2333/5/$20.00

2333

© 2003 American Institute of Physics


2334

Kucur et al.

J. Chem. Phys., Vol. 119, No. 4, 22 July 2003

FIG. 3. Oxidation voltammograms of monodisperse CdSe nanocrystals of different sizes (a⫽3.23 nm, b⫽3.48 nm, c⫽3.73, d⫽3.80 nm) adsorbed on a gold disk microelectrode 共400 ␮m radius兲 in acetonitrile⫹0.1 M TBAPF6 at 20 mV/s 共room temperature兲. E p indicates the peak potentials, E ons the location of the onset potentials.

FIG. 1. TEM micrograph of CdSe nanocrystals. Size: 3.73 nm.

Acetonitrile 共Aldrich兲 was dried over calcium hydride 共Avocado兲. Tetrabutylammoniumhexafluorophosphate (TBAPF6 ) was purchased from Merck and used as received. A 100 mM solution of TBAPF6 共supporting electrolyte兲 in acetonitrile was prepared within a drybox and used as electrolyte solution for all electrochemical measurements.

deposition. The electrodes were etched using reactive ion equipment. CdSe nanocrystals were dispersed in chloroform within nitrogen atmosphere. Microelectrodes were dip-coated with this nanocrystal dispersion and dried for 2 h at 120 °C. This process led to adsorption of nanocrystals on the electrode surface. Subsequently, the nanocrystal-coated electrodes were transferred to the electrochemical cell under Schlenk conditions. C. Apparatus

B. Electrode preparation

Gold disk microelectrodes with diameters of 800 ␮m were used as working electrodes. They were produced using thin-film technology. A Ti–Au–Ti 共100–200–100 nm兲 metallization deposited by means of high vacuum evaporation was patterned with a lift-off process and insulated with 1 ␮m silicon nitride by means of plasma enhanced chemical vapor

Voltammetric measurements were performed within a home-made electrochemical inert gas cell. The cell was mounted within a Faraday cage. Electrochemical instrumentation consisted of a HEKA PG 310 potentiostat/galvanostat. Potentials were recorded versus a home-made Ag兩0.01 M AgNO3 ⫹0.1 M TBAPF6 acetonitrile reference electrode at room temperature. Transmission electron microscopy 共TEM兲 measurements were carried out on a LEO 912 Omega operating at 120 kV. UV/vis and photoluminescence spectroscopy was conducted on a J&M TIDAS diode-array spectrometer. A J&M FL3095 monochromator was used as monochromatic light source. A halogen/deuterium combination was used as light source for the absorption spectra. III. RESULTS

All electrochemical and spectroscopic measurements were performed using spherical CdSe nanocrystals of four sizes 共diameters兲: 3.23, 3.48, 3.73, and 3.80 nm. Particle sizes were approximated from the ‘‘effective band gap’’ values E g from the absorption edges and calculated according to:18 FIG. 2. Cyclic voltammogram of monodisperse CdSe nanocrystals 共3.73 nm diameter兲 adsorbed on a gold disk microelectrode 共400 ␮m radius兲 in acetonitrile⫹0.1 M TBAPF6 at 20 mV/s at room temperature. Dotted line: voltammogram of bare gold electrode.

a⫽

冑

冉

冊

h2 1 1 ⫹ , 8⌬E g m e m h

共1兲

where a is the crystal diameter, m e and m h are the effective electron and hole masses (m e ⫽0.13m 0 ,m h ⫽0.44m 0 for



Determination of quantum confinement in CdSe nanocrystals

2335

In order to estimate the ionization potential I p and the electron affinity E.A. of the nanocrystals from their oxidation and reduction waves it is necessary to relate the electrochemical potentials to the vacuum level. Such an estimation has been done for electrochemical measurements of conducting polymers14,15 and tantalum oxide, -nitride, respectively.21 It is expected that I p ⫽⫺ 共 E ox⫹4.14兲 eV,

共2兲

E.A.⫽⫺ 共 E red⫹4.14兲 eV,

共3兲

where E ox and E red are the onset potentials of the oxidation and reduction waves relative to an Ag兩0.01 M AgNO3 reference electrode.16 The value of 4.14 represents the difference between the vacuum level potential of the normal hydrogen electrode 共NHE⫽4.44 eV兲 and the potential of the Ag兩 AgNO3 -electrode. Table I outlines the peak and onset potentials for the oxidation and reduction of adsorbed nanocrystals. The values were determined from the cyclic voltammograms displayed in Figs. 3 and 4. Figure 3 depicts the peak potentials E p and the onset potentials E ons were obtained. The difference ⌬E ⫽E ox⫺E red should be similar to the effective band gap of the semiconducting particles.22 The ⌬E values from Table I show that the utilization of onset potentials does not make sense for this type of measurement, because the band gap values calculated from the onset potentials are smaller than the bulk band gap of CdSe 共1.74 eV兲.

FIG. 4. Reduction voltammograms of monodisperse CdSe nanocrystals of different sizes (a⫽3.23 nm, b⫽3.48 nm, c⫽3.73, d⫽3.80 nm) adsorbed on a gold disk microelectrode 共400 ␮m radius兲 in acetonitrile⫹0.1 M TBAPF6 at 20 mV/s 共room temperature兲.

bulk CdSe, m 0 as electron rest mass兲, ⌬E g the shift of the spectroscopic band gap relative to the bulk band gap19 (⌬E g ⫽E g ⫺1.74 eV), and h is Planck’s constant. Additional transmission electron micrographs revealed that the standard deviation of the monodispersity was better than 5%. Figure 1 shows a typical transmission electron micrograph of CdSe nanocrystals. A. Electrochemical measurements

B. Spectroscopic measurements

Figure 2 shows the cyclic voltammogram of monodisperse CdSe nanocrystals 共3.73 nm diameter兲 adsorbed on a gold disk microelectrode 共400 ␮m radius兲 in acetonitrile ⫹0.1 M TBAPF6 at 20 mV/s. The voltammogram shows an oxidation peak—i.e., the ionization potential—at approximately 1.4 V versus Ag兩0.01 M AgNO3 ⫹0.1 M TBAPF6 and a less prominent reduction wave—i.e., the electron affinity—at about ⫺0.75 V. Compared to recent measurements of colloidal semiconductors,20 these waves are very prominent and can clearly be assigned to the oxidation and reduction potentials. The scan speed was increased up to 1 V/s, but no shift of the characteristic potentials was found. For clarity, the oxidation and reduction waves were magnified and displayed, for all four sizes of nanocrystals together, in separate figures. Figure 3 depicts the oxidation waves for nanocrystals ranging from 3.2 to 3.8 nm in diameter; Fig. 4 the corresponding reduction waves. The experimental conditions were as above.

In cases where the reduction potentials of semiconducting materials cannot be determined easily, it has been common practice to estimate E a by subtraction of the optically determined band gap E g from the I p value determined either electrochemically or by other methods.23,24 E g can be estimated from the absorption onsets of the material in question. Figure 5 shows the absorption spectra of the CdSe nanocrystal colloids used in this study. All spectroscopic measurements were performed in chloroform and the concentration of the colloids was adjusted to an optical density of approximately 0.05. The E g values, determined from Fig. 5, are outlined in Table II. Table II also contains the ⌬E values, estimated from the electrochemical oxidation and reduction potentials for comparison. A third possibility to estimate the band gap of semiconducting nanocrystals 共direct semiconductors兲 includes the analysis of photoluminescence emission spectra. The photo-

TABLE I. Oxidation and reduction potentials of CdSe nanocrystals by means of cyclic voltammetry. Onset potentials E ons and peak potentials E p were determined as schematically shown in Fig. 3. ⌬E is calculated according to ⌬E⫽E ox⫺E red . 3.23 nm

E ox 共V兲 E red 共V兲 ⌬E 共V兲

3.48 nm

3.73 nm

3.80 nm

Onset

Peak

Onset

Peak

Onset

Peak

Onset

Peak

1.13 ⫺0.38 1.51

1.46 ⫺0.64 2.10

1.12 ⫺0.56 1.68

1.37 ⫺0.66 2.03

0.87 ⫺0.56 1.43

1.35 ⫺0.64 1.99

0.72 ⫺0.65 1.37

1.17 ⫺0.77 1.94


2336

Kucur et al.


FIG. 5. Absorption spectra of CdSe nanocrystals of four different sizes. E g indicates the band gap values determined from these spectra and listed in Table II.

luminescence spectra of the CdSe nanocrystal colloids are depicted in Fig. 6, and the effective band gap values E ⬘g determined from these spectra are included in Table II. Although the quantum confinement is seen very well in these values, a quantitative estimation of the band gap is difficult as the photoluminescence energy depends strongly on the nanocrystals surface.25 Furthermore, the determination of absolute band positions is not possible at all. IV. DISCUSSION

First, it has to be stated that cyclic voltammetry is a well suited method to measure the absolute vacuum level of the ionization potential I p and the electron affinity E.A. of semiconductor nanocrystals. It was found that the oxidation potential, which corresponds to I p , is very prominent, whereas the reduction wave, corresponding to E.A. emerges with a much lower current. Quantum confinement in nanocrystals of different sizes and the resulting shift of I p and E.A. can clearly be seen. Herewith, the electrochemical results are in total agreement with the theoretical expectations12,13,22 and the spectroscopic data. It was found that the oxidation and reduction potentials of the nanocrystal films do not depend on the scan speed within the examined range. This means that thermodynamic instead of kinetic values were measured, which is in full agreement with the above-noted findings. The comparison between the onset values and the peak values 共displayed in Table I兲 reveals that onset values are a

FIG. 6. Photoluminescence spectra of CdSe nanocrystals of four different sizes: a⫽3.23 nm, b⫽3.48 nm, c⫽3.73, d⫽3.80 nm. Excitation at 366 nm for all samples.

poor approximation 共band gap energies calculated from onset values are below the bulk band gap energy E g ⫽1.74 eV) in contrast to conducting polymers. The band gap energies calculated from the peak potentials are in almost full agreement with the E g values determined from the absorption spectra 共cf. Table II兲. Thus it can be concluded that I p and E a can be estimated well from the anodic, respectively, cathodic peak potentials with this method. Figure 7 depicts schematically the electrochemically determined I p and E a values on an electrochemical scale and the vacuum level calculated from Eqs. 共2兲 and 共3兲. It was found that the ⌬E values are in good agreement with the optical measurements. The general tendency of the I p and E.A. values is in good agreement with the expectations based on the quantum confinement effect and predicted theoretically.22

TABLE II. Estimated band gap energies for different CdSe nanocrystal colloids by three methods: absorption onset E g , photoluminescence maximum E g⬘ , and electrochemical determination ⌬E.

E g 共eV兲 E ⬘g 共eV兲 ⌬E 共V兲

3.23 nm

3.48 nm

3.73 nm

3.80 nm

2.10 2.17 2.10

2.05 2.13 2.03

2.01 2.08 1.99

2.00 2.05 1.94

FIG. 7. Schematic diagram of the electrochemically determined ionization potentials I p and electron affinities E.A. of four different sizes of CdSe nanocrystals. Dashed arrows: Approach of the confined I p and E a values to the bulk band edges with growing particle size 共CdSe bulk band gap ⫽1.74 eV兲.



V. CONCLUSIONS

In summary, an electrochemical method for the measurement of the ionization potential I p and the electron affinity E.A. in CdSe nanocrystals has been suggested. It was shown that the spectroscopically observed and theoretically expected quantum confinement with decreasing particle size can be detected. This method is easier than, e.g., the ultraviolet photoelectron spectroscopy and includes the additional advantage of compatibility to analogous measurements with conducting polymers. This is of special interest for nanocrystal/conducting polymer composite materials used for novel light-emitting or photovoltaic devices and polymer microelectronics. 1

T. S. Ahmadi, Z. L. Wang, T. C. Green, A. Henglein, and M. A. Elsayed, Science 272, 1924 共1996兲. 2 W. U. Huynh, J. J. Dittmer, and A. P. Alivisatos, Science 295, 2425 共2002兲. 3 W. U. Huynh, X. Peng, and A. P. Alivisatos, Adv. Mater. 11, 923 共1999兲. 4 D. S. Ginger, N. Greenham, X. Peng, and A. P. Alivisatos, Synth. Met. 84, 545 共1997兲. 5 J. Lee, V. C. Sundar, J. R. Heine, M. G. Bawendi, and K. F. Jensen, Adv. Mater. 12, 1102 共2000兲. 6 N. Tessler, V. Medvedev, M. Kazes, S. Kann, and U. Banin, Science 295, 1506 共2002兲. 7 B. O. Dabbousi, M. G. Bawendi, O. Onitsuka, and M. F. Rubner, Appl. Phys. Lett. 66, 1316 共1995兲.

Determination of quantum confinement in CdSe nanocrystals

2337

V. L. Colvin, M. C. Schlamp, and A. P. Alivisatos, Nature 共London兲 370, 354 共1994兲. 9 I. Willner, F. Patolsky, and Wasserman, J. Angew. Chem. Int. Ed. 40, 1861 共2001兲. 10 M. Bruchez, Jr., M. Moronne, P. Gin, S. Weiss, and A. P. Alivisatos, Science 281, 2013 共1998兲. 11 W. C. W. Chan and S. Nie, Science 281, 2016 共1998兲. 12 L. E. Brus, J. Chem. Phys. 80, 4403 共1984兲. 13 A. Zunger, Mater. Res. Bull. 23, 35 共1998兲. 14 J. L. Bre´das, R. Silbey, D. S. Boudreaux, and R. R. Chance, J. Am. Chem. Soc. 105, 6555 共1983兲. 15 S. Janietz, D. D. C. Bradley, M. Grell, C. Giebeler, M. Inbasekaran, and E. P. Woo, Appl. Phys. Lett. 73, 2453 共1998兲. 16 S. Trasatti, in Electrified Interfaces in Physics, Chemistry and Biology, edited by R. Guidelli 共Kluwer Academic, Dordrecht, 1992兲, pp. 229–244. 17 T. Nann and J. Riegler, Chem. Eur. J. 8, 4791 共2002兲. 18 J. Hambrock, A. Birkner, and R. A. Fischer, J. Mater. Chem. 11, 3197 共2001兲. 19 Handbook of Chemistry and Physics, 76th ed., edited by D. R. Lide 共CRC, Boca Raton, FL, 1995兲. 20 S. K. Haram, B. M. Quinn, and A. J. Bard, J. Am. Chem. Soc. 123, 8860 共2001兲. 21 W.-J. Chun, A. Ishikawa, H. Fujisawa, T. Takata, J. N. Kondo, M. Hara, M. Kawai, Y. Matsumoto, and K. Domen, J. Phys. Chem. B 107, 1798 共2003兲. 22 L. E. Brus, J. Chem. Phys. 79, 5566 共1983兲. 23 W. Rieß, Polym. Adv. Technol. 8, 381 共1997兲. 24 D. O’Brien, A. Bleyer, D. G. Lidzey, D. D. C. Bradley, and T. Tsutsui, J. Appl. Phys. 82, 2662 共1997兲. 25 H. Fu and A. Zunger, Phys. Rev. B 56, 1496 共1997兲. 8
