Photoluminescence response of colloidal quantum dots on VO2 film across metal to insulator transition
© Kuznetsov et al.; licensee Springer. 2014
Received: 17 September 2014
Accepted: 4 November 2014
Published: 13 November 2014
We have proposed a method to probe metal to insulator transition in VO2 measuring photoluminescence response of colloidal quantum dots deposited on the VO2 film. In addition to linear luminescence intensity decrease with temperature that is well known for quantum dots, temperature ranges with enhanced photoluminescence changes have been found during phase transition in the oxide. Corresponding temperature derived from luminescence dependence on temperature closely correlates with that from resistance measurement during heating. The supporting reflectance data point out that photoluminescence response mimics a reflectance change in VO2 across metal to insulator transition. Time-resolved photoluminescence study did not reveal any significant change of luminescence lifetime of deposited quantum dots under metal to insulator transition. It is a strong argument in favor of the proposed explanation based on the reflectance data.
71.30. + h; 73.21.La; 78.47.jd
The so-called correlated oxides attract much attention due to their prospective applications in electronic and optoelectronic devices (see  for review). Vanadium dioxide is the most known among them according to its near room temperature metal-insulator transition (MIT) at approximately 67°C. Nowadays, a tendency arises to couple VO2 with other materials that will promote a search for new application fields. For example, phase transition in VO2 provides a tunability of localized plasmon resonance in metallic sub-wavelength structures . The incorporation of photoluminescence (PL) ions into VO2 could result in a novel photonic material capable of simultaneous optical switching and amplification . In the bilayer structure of ZnO/VO2, MIT in VO2 was shown to control a photoluminescence response of zinc oxide . To our opinion, it is of interest to study another combination of VO2 with highly luminescent fluorophores such as colloidal quantum dots (QDs). In contrast to , a deposition of self-assembled QD layer on a VO2 film does not require a high-temperature treatment; therefore, the intrinsic properties of vanadium dioxide will be unchanged. In addition, due to QD's small diameter less than 10 nm and amorphous layer structure, the deposited coating is tightly bound with coarse polycrystalline VO2 film. This circumstance may stimulate more strong interaction between QDs and VO2. In this letter, we report on first observation the PL response of core/shell CdSe/ZnS layer deposited on the VO2 film when the latter is passing MIT. Based on well-known intrinsic properties of these QDs, some features in PL response have been revealed in the course of phase transition of the underlying VO2 film. Using the unified mathematical treatment of electrical and optical temperature dependencies, we derive and compare characteristic parameters for the system studied. From the comparison, the leading role of MIT in the oxide component of hybrid structure is evident. Besides, our finding provides the useful method to probe MIT in the correlated oxides.
Vanadium dioxide films were grown by reactive DC magnetron sputtering on polished sitall (Russian brand of glass ceramics) substrates. Deposition was carried out with AJA International sputtering system. First, the substrates were cleaned with RF etching in Ar. Before deposition, pure vanadium target was cleaned in optimal DC regime at 200 W in a mixture of Ar and O2 (flow rates of 14 and 2 sccm, respectively, total pressure of 5 mTorr) while the target was closed by a shutter. After the above preliminary procedures, the shutter was turned off and a deposition was carried out for 20 min on the sitall substrate at room temperature. Immediately after deposition, a film was annealed at 520°C for 40 min in pure O2 at a pressure of 10 mTorr. These conditions provide the deposition of vanadium dioxide layer with approximately 150 nm in thickness. Fabricated VO2 films were polycrystalline with rather wide dispersion in grain sizes. Mean grain dimension was near 90 nm as revealed by AFM measurement in contact mode.
In the second stage of the present study, the VO2 films were decorated with colloidal core/shell quantum dots by spin-coating. CdSe/ZnS dots were obtained from Evident Technologies. They consist of CdSe core with diameter of ≈ 5 nm and ZnS shell ≈ 2 nm thick. These QDs are characterized by high quantum efficiency PL of 90% at a peak wavelength of approximately 630 nm and high stability against moderate heating up to approximately 100°C. QDs were spin-coated from a dilute solution in toluene on the VO2 film at approximately 2000 rpm. After drying in the air at room temperature, the obtained layers were baked at 100°C in the air for 1 h. This procedure served to minimize an irreversible change of photoluminescence response in further heating/cooling cycles. QD layer thickness was ≈ 50 nm as revealed by a profilometer. PL measurements were carried out under UV excitation in narrow band centered at 385 nm (Nichia UV-LED) with excitation power density of approximately 1 mW/cm2. PL spectra were recorded by grating spectrometer with TE-cooled CCD camera and were corrected for spectral response. For temperature-dependent PL measurement, a sample was mounted in homemade temperature cell. A temperature change rate under heating/cooling was not constant, but its value did not exceed 10°C/min. According to , there is no significant shift in phase transition parameters of VO2 at this rate. It should be noted that a temperature was recorded by a thermopile aligned to the outer surface of the VO2 film. We estimate an uncertainty in absolute temperature of a sample which does not exceed ±1°C. PL decay measurements were performed on epifluorescence microscope. The excitation source was a pulsed laser diode operating at 405 nm with 300 ps pulse duration and 4 MHz repetition rate. QD luminescence is collected using the same microscope objective with band-pass filter and detected by avalanche photodiodes operated in Geiger mode. Lastly, transient emission dynamics are analyzed by a time-correlated single-photon counting module. The overall temporal resolution was better than 100 ps.
Results and discussion
A comparison of characteristic temperatures obtained by different methods
MIT under heating, °C
Transition width (FWHM), °C
MIT under cooling, °C
Transition width (FWHM), °C
Hysteresis loop width, °C
Reflectance at 385 nm
A qualitative observation should be added for further discussion. If a VO2 film with strong MIT was annealed at 300°C in the air before QD deposition, then the above features in PL response did not appear. At the same time, the annealing dramatically restricted MIT resistance change in that film due to VO2 conversion into higher oxides. Therefore, we can refer observed PL features (with their proper parameters) to MIT origin in VO2.
As can be seen in Figure 6, QDs on the VO2 film exhibit higher PL decay rate than that on the dielectric substrate irrespective of temperature. Though this feature is out of scope of the present study, let us touch upon the question. It is noted in the ‘Methods’ section that QD layers were deposited atop substrates by spin-coating from toluene solution followed by baking at approximately 100°C for 1 h in the air. The latter treatment is aimed to decrease a degradation of QD's PL properties in the course of subsequent heating/cooling measurement cycles. This annealing of QDs deposited on VO2 led to more pronounced PL decrease (more than twofold) against that for simple drying at RT. It should be stressed that there is no significant PL decrease when the same preparation was done on the dielectric substrate. We can assume that PL efficiency of QDs on VO2 is quenched due to well-known catalytic activity of vanadium oxides.
Let us discuss characteristic temperatures obtained by three methods. As can be seen in Table 1, satisfactory coincidence between characteristic temperatures exists for insulating to metallic phase change under heating (within our temperature uncertainty of ±1°C). For reverse phase transition (cooling), there is a prominent discrepancy between the temperature derived from electrical measurement and those for both optical characterization methods.
Only few works are devoted to combined experimental study of MIT phenomenon including electrical and optical characterization methods. The work  presents simultaneous optical and electrical measurements of MIT in a two-dimensional nanostructured VO2 film and shows that there is an offset between the effective optical and electrical switching temperatures. The authors proposed that this offset arises because the hysteresis in optical transmittance is determined by the relative fraction of grains in semiconducting state, while the hysteresis in electrical resistance is governed by the evolution of a continuous percolation path. It is obvious that this explanation is not valid for thick polycrystalline films.
The authors of  observed the difference in MIT temperatures probing by electrical (resistance) and optical (reflectance in IR) measurements. However, the cited authors analyzed the results only for heating branch. It was found that switching temperature derived from resistance data has significant shift against optically derived temperature. To clarify the origin of observed difference, the authors revised their electrical measurements in terms of conductance instead of resistance and found much less discrepancy. They concluded that conductance instead of resistance is a more suitable characteristic in comparison with optical data. Note that the same mathematical treatment (see above) for extraction of characteristic temperature from electrical measurement was used in the cited work. Then, it is simple to show from relation σ = R-1 that d(Log σ)/dT = -d(Log R)/dT, where σ is the conductance and R is the resistance. Therefore, such substitution cannot change the final result. Our detailed consideration of the cited work is aimed to verify the validity of resistance-based approach used in our study.
Thus, we observe the nonequivalence of MIT trajectories under heating and cooling that is manifested by the difference between characteristic temperatures derived from electrical and optical data. We assume that more systematic and sophisticated experimental studies are needed to clarify an underlying mechanism.
We have probed the phase transition in the vanadium dioxide film decorated with colloidal quantum dots measuring a temperature dependence of PL response. Characteristic temperatures derived from PL measurement are compared with those obtained by standard conductivity method. The reflectance dependence on temperature taken at PL excitation wavelength points out that PL mimics a reflectance change in VO2 across MIT. No evidence was observed in favor of strong interaction between QDs and VO2 caused by MIT.
The authors acknowledge the financial support from Ministry of Education and Science of Russian Federation under contract no. 3.757.2014/K and partial support from Program for Strategic Development of Petrozavodsk State University. SNK thanks Mr. A.V. Lemetti for technical assistance.
- Yang Z, Ko C, Ramanathan S: Oxide electronics utilizing ultrafast metal-insulator transitions. Annu Rev Mater Res 2011, 41: 337–367. 10.1146/annurev-matsci-062910-100347View ArticleGoogle Scholar
- Earl SK, James TD, Davis TJ, McCallum JC, Marvel RE, Haglund RF Jr, Roberts A: Tunable optical antennas enabled by the phase transition in vanadium dioxide. Optics Express 2013, 21: 27503–27508. 10.1364/OE.21.027503View ArticleGoogle Scholar
- Lim H, Stavrias N, Johnson BC, Marvel RE, Haglund RF, McCallum JC: Optical switching and photoluminescence in erbium-implanted vanadium dioxide thin films. J Appl Phys 2014, 115: 093107. 10.1063/1.4867481View ArticleGoogle Scholar
- Srivastava A, Herng TS, Saha S, Nina B, Annadi A, Naomi N, Liu ZQ, Dhar S, Ariando , Ding J, Venkatesan T: Coherently coupled ZnO and VO2 interface studied by photoluminescence and electrical transport across a phase transition. Appl Phys Lett 2012, 100: 241907. 10.1063/1.4729387View ArticleGoogle Scholar
- Brassard D, Fourmaux S, Jean-Jacques M, Kieffer JC, El Khakani MA: Grain size effect on the semiconductor–metal phase transition characteristics of magnetron-sputtered VO2 thin films. Appl Phys Lett 2005, 87: 051910. 10.1063/1.2001139View ArticleGoogle Scholar
- Appavoo K, Lei DY, Sonnefraud Y, Wang B, Pantelides ST, Maier SA, Haglund RF Jr: Role of defects in the phase transition of VO2 nanoparticles probed by plasmon resonance spectroscopy. Nano Lett 2012, 12: 780–786. 10.1021/nl203782yView ArticleGoogle Scholar
- Gurvitch M, Luryi S, Polyakov A, Shabalov A: Nonhysteretic behavior inside the hysteresis loop of VO2 and its possible application in infrared imaging. J Appl Phys 2009, 106: 104504. 10.1063/1.3243286View ArticleGoogle Scholar
- Pugh-Thomas D, Walsh BM, Gupta MC: CdSe(ZnS) nanocomposite luminescent high temperature sensor. Nanotechnology 2011, 22: 185503. 7pp 7pp 10.1088/0957-4484/22/18/185503View ArticleGoogle Scholar
- Kana Kana JB, Ndjaka JM, Vignaud G, Gibaud A, Maaza M: Thermally tunable optical constants of vanadium dioxide thin films measured by spectroscopic ellipsometry. Optics Communications 2011, 284: 807–812. 10.1016/j.optcom.2010.10.009View ArticleGoogle Scholar
- Jasieniak J, Smith L, van Embden J, Mulvaney P, Califano M: Re-examination of the size-dependent absorption properties of CdSe quantum dots. J Phys Chem C 2009, 113: 19468–19474. 10.1021/jp906827mView ArticleGoogle Scholar
- Rozen J, Lopez R, Haglund RF Jr, Feldman LC: Two-dimensional current percolation in nanocrystalline vanadium dioxide films. Appl Phys Lett 2006, 88: 081902. 10.1063/1.2175490View ArticleGoogle Scholar
- Ruzmetov D, Zawilski KT, Senanayake SD, Narayanamurti V, Ramanathan S: Infrared reflectance and photoemission spectroscopy studies across the phase transition boundary in thin film vanadium dioxide. J Phys:Condens Matter 2008, 20: 465204. 10.1088/0953-8984/20/46/465204Google Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited.