Advertisement

Applied Physics A

, 122:412 | Cite as

Photoluminescence excitation of lithium fluoride films by surface plasmon resonance in Kretschmann configuration

  • Jiří Bulíř
  • Tomáš Zikmund
  • Michal Novotný
  • Ján Lančok
  • Ladislav Fekete
  • Libor Juha
Article
  • 187 Downloads
Part of the following topical collections:
  1. Emerging trends in photo-excitations and promising new laser ablation technologies

Abstract

We report on excitation of the photoluminescence of lithium fluoride by means of the surface plasmon resonance of Al layer. Advantage of this method is high efficiency of the excitation, which is applicable to ultra-thin films. P-polarized UV diode laser light is coupled to the surface plasmon resonance using a fused silica prism in Kretschmann configuration. The angular dependence of the reflected intensity is measured using a theta–2theta goniometer. The surface plasmon at resonance condition induces photoluminescence in the adjacent lithium fluoride layer. The fluoride layers were deposited on Al-coated fused silica substrates by electron beam evaporation. For the experiment, we prepared several samples with thickness ranging from 20 to 71 nm. We studied the effect of the luminescence enhancement by the surface plasmon resonance effect. Strong quenching effect was observed in the thinnest LiF layer. Influence of X-ray irradiation on the photoluminescence was studied.

Keywords

Surface Plasmon Resonance Surface Plasmon Polariton Color Center Surface Plasmon Resonance Peak Lithium Fluoride 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

1 Introduction

Alkali halides are a group of material with unique properties. Among them, the lithium fluoride (LiF) occupies a special place because of its unique properties such as low refractive index and very large bandgap of 13.6 eV allowing most extreme UV transmission beyond wavelength of 120 nm. Lithium fluoride can be used as a phosphor, which is highly sensitive for ionizing radiation, especially when it is doped with a proper activator. It is known for phenomenon of coloring of compounds due to color centers (CC). The color center is a crystallographic defect, in which one or more electrons are trapped in an anionic vacancy [1]. The primary color center (F) has strong absorption at 250 nm. Till now, the photoluminescence was not detected originating from this CC. The individual color centers can be aggregated together forming aggregate of two (F2) or three (F3) CCs. Moreover, the CC can be ionized negatively or positively if one electron is attached to or removed from the CC, respectively. The absorption of F2 and F3 + CC is almost overlapping around 450 nm. This absorption varies in a large range depending on the CC concentration [2, 3]. Under the optical pumping at this wavelength, they exhibit efficient emissions peaking around 670 and 540 nm for F2 and F3 +, respectively [1, 4]. These defects can be produced by the irradiation of the LiF material with ionizing radiations such as electrons, ions, extreme ultraviolet, X-rays, and gamma rays. The coloration of the LiF occurs in the surface layer, which is limited by the penetration depth of the radiation in the LiF material [3]. There is a high interest about the color centers because of their potential applications [5]. They can be used for optical fluorescence scanning microscopies, optical microcavities, X-ray imaging, proton beam diagnostics and dose mapping, and laser active electronic states, which are characterized by wide tunability and good stability at room temperature.

The surface plasmon resonance (SPR) has attracted substantial interest in many applications such as biosensors [6] and bioimaging [7], surface-enhanced Raman scattering, solar cells, and others. SPR is characterized as collective oscillations of free carriers of charge coupled to electromagnetic waves as surface plasmon polariton (SPP). This coupling can be realized as an interface between dielectrics and metal layer in Kretschmann [8] or Otto configuration [9] or via metallic grating. At the resonant condition, there is a significant enhancement of the electric field of the incident light at the metal surface by several factors of ten. This effect can be used for significant enhancement of photoluminescence in the adjacent fluorescent layer due to the surface plasmon-coupled emission (SPCE) [7, 10, 11]. This effect enables efficient photoluminescence excitation in very thin layers. Moreover, coupling of the surface plasmon to Fabry–Perot resonance was investigated [12, 13]. In that case, directionality of the photoluminescence emission can be tailored as it is shown in [13] where the Fabry–Perot resonator is accomplished using metal–dielectric–metal structure. Selection of suitable plasmonic metal plays an important role. In the ultraviolet region, aluminum is a good choice because its dielectric constant possesses negative real part and small imaginary part. Effect of the Al layer thickness and the adjacent fluorescent layer thickness on SPR was studied in Refs. [7, 10, 14]. In these works, it is shown that shape and amplitude of the electric field intensity peak correlate with the SPR dip in the reflectance spectrum. In this manner, it corresponds to the fluorescence intensity enhancement.

We report on study of the photoluminescence excitation using the surface plasmon resonance coupling on Al layer in the Kretschmann configuration. The aim of this study was to enhance efficiency of the luminescence excitation allowing to study very thin luminescent layers.

2 Reflectance calculation

Transfer matrix method generalized for absorbing materials was used for the reflectance calculation. The reflectance of m plan-parallel layers is calculated using the characteristic matrix of the system and generalized Fresnell coefficients [15].

Figure 1 shows dependence of the Al layer reflectance on the layer thickness and the incident angle for the wavelength of 403 nm in Kretschmann configuration. It is considered that the light polarization is parallel to the plane of incidence (p-polarization). There is a threshold of total reflection seen at 40.5° as the dark–light edge on the upper part of the image where the Al thickness goes to zero. The attenuated total reflection is seen at the right part of the image for the nonzero thickness of the Al layer. In this zone, there is seen a significant narrow dip of the reflectance on approx. 45° caused by the maximal SPR. The strongest coupling of SPR is characterized by deep and narrow minimum in the reflectance spectrum. The optimal Al layer thickness of 16.4 nm with the strongest SPR coupling is marked in the Fig. 1 by dashed line. Empirically, it can be observed that the width of the reflectance dip is inversely proportional to the electric field intensity on the metal surface [10, 16]. The very narrow peak of the SPR in the UV spectral range makes Al appropriate material for plasmonic applications in this area.
Fig. 1

Simulated p-polarized reflectance of the Al layer in Kretschmann configuration. The color map depicts dependence of reflectance on incident angle and Al layer thickness

There is strong dependence of SPR coupling on the adjacent dielectric layer on the Al surface. This effect is related to SPP dispersion equation representing the conservation of wave vector and frequency at the metal–dielectric interface. In the resonant condition, the wave vector of the incident light in the plane of the metal–dielectric interface matches the wave vector of the surface plasmon wave in the metallic film [17]. The angular dependence of the SPR on refractive index of the surrounding media can be derived from that relation, e.g., depicted in [17]. The thickness of the adjacent thin dielectric layer possesses effectively similar influence. Figure 2 shows simulation of the LiF layer on the Al layer with the optimized thickness of 16.4 nm. The color map shows dependence of the p-polarized reflectance on the LiF layer thickness and the incident angle. With increase in the LiF layer thickness, there is significant shift of the reflectance dip seen from 45° to the higher angle saturating at 80°. Simultaneously, this shift is associated with broadening of the dip. With respect to the dip broadening and the measurement possibilities, we chose the optimal thickness of the LiF layer in the range between 0 and 70 nm. For the LiF layer thickness exceeding approximately 170 nm, very narrow dip occurs that is caused by interference effect of coupled waveguide-plasmon resonance [18].
Fig. 2

Simulated p-polarized reflectance of LiF/Al structure in Kretschmann configuration. The color map depicts dependence of reflectance on incident angle and LiF layer thickness. The thickness of the Al layer is considered 16.4 nm

3 Experimental

The Al layers were deposited in a vacuum chamber by a radiofrequency (13.56 MHz) magnetron sputtering of an Al target. The r.f. power was set to 400 W. The vacuum chamber was evacuated to ultimate pressure of 2 × 10−5 Pa before each deposition. The Ar gas pressure was adjusted to 0.5 Pa by selecting the Ar gas flow to 20 sccm and by subsequent tuning of the throttle valve between the deposition chamber and the vacuum turbomolecular pump. The Al target was pre-sputtered for 120 s prior the deposition in order to clean the target surface from native oxide. The Al layers were grown on fused silica substrates with dimensions of 12 × 14 mm2. The substrates were carefully cleaned and annealed at 200 °C for 2 h prior the deposition.

The LiF films were deposited by an e-flux Mini Electron Beam Evaporator (Tectra GmbH) equipped with a molybdenum crucible with the internal diameter of 6 mm. The crucible was filled by sintered LiF pellets of 99.95 % purity. The electron accelerating voltage was set to 1 kV. The emission current was stabilized at 10 mA by a feedback control of a filament current between 6.8 and 7.0 Å. The deposition rate was 0.16 nm/s. The ultimate pressure of the vacuum chamber was 5 × 10−6 Pa, but during the deposition process, it increased to about 3 × 10−5 Pa. The substrate was placed at a distance of 12 cm from the crucible. The substrate was heated by a bulb reflector heater to temperature about 150 and 300 °C during the deposition process. The LiF layers were deposited on Al-coated fused silica, bare fused silica, and Si substrates. The LiF-coated Si substrates were used for X-ray diffraction (XRD), scanning electron microscopy (SEM), and optical measurements. The LiF-coated bare fused silica substrate was used as a reference sample for the photoluminescence measurement.

Optical constants of the deposited Al and LiF layers were analyzed by means of spectral ellipsometer (J.A. Woollam Co., M2000) in the wavelength range from 245 to 1000 nm. The optical constants were obtained from the ellipsometric data by fitting with an appropriate model. For LiF, we used the Cauchy dispersion model; for the Al layer, we used a dispersion model composed of several Lorentz oscillators. The Al layer thickness was obtained by a profilometer (Tencor P-06). The samples were characterized by X-ray diffractometer X’Pert with Co Kα radiation. The crystallite size was estimated using Scherrer’s equation from the measured diffraction peaks after subtraction of instrumental broadening. The same X-ray source was used for sample’s irradiation (photon energy 6.9 keV, wavelength 0.1789 nm) for 60 min. Composition of the LiF layer was characterized by X-ray photoemission spectroscopy (XPS)—using NanoESCA instrument. The spectra were acquired by Al Kα excitation.

Photoluminescence excitation was performed using surface plasmon resonance coupling in Kretschmann configuration. The measurement setup is shown in Fig. 3. The measured sample consists of the LiF/Al layer deposited on fused silica substrate. The sample is attached to a fused silica prism using fluorescence-free immersion oil in order to avoid unwanted reflection on the prism–substrate interface. For the angular-dependent measurement, we used a theta–2theta goniometer. The prism with the sample is fixed to the rotating table. For the measurement of angular dependency of reflectance, we used a Si photodiode detector mounted on the rotating arm of the goniometer. Rotation of the arm and the table is driven by stepper motors, which are controlled by a computer. For the reflectance measurement, we used the optically pumped semiconductor laser (Coherent OBIS 405LX) with the adjustable output power set to 10 mW and the wavelength of 403 nm. The polarization of laser beam was set parallel with the plane of incidence of the sample (p-polarization). The measured signal was normalized to the signal of the direct laser beam. The raw measurement data were corrected for losses caused by reflected light on prism sides. The goniometer angle was converted to internal incident angle in the prism applying the Snell law to the refracted light on the prism sides.
Fig. 3

Experimental setup of angular measurement of reflectance and photoluminescence

For the excitation of the photoluminescence, we used both: the above-mentioned laser with the output power increased to 50 mW and the diode laser with the constant output power of 500 mW working at the wavelength of 445 nm. The photoluminescence was detected using a miniature spectrophotometer, Ocean Optics, USB4000, which was attached to the experimental setup using an optical fiber (core diameter 600 µm) and a focusing lens (focus length 8 mm) as a light collector. The light collector was fixed to the rotating table in front of the measured sample. Edge long-pass optical filter (optical density, OD4) with cut-on wavelength of 425 or 475 nm was inserted between the collecting lens and the optical fiber for blocking the excitation wavelength of the 403 or 445 nm laser, respectively. The measured data were corrected using a spectral-dependent, sensitivity curve of the spectrophotometer. This sensitivity curve was obtained from measurement of a calibrated tungsten lamp (20 W) with known spectral irradiance at the distance of 500 mm.

4 Results

The high-quality Al layer is important prerequisite of the efficient coupling of the light to the surface plasmon resonance. Plasmonic applications require metal with low optical losses. This behavior can be achieved for the metal material with negative real permittivity while the low losses are represented by near-to-zero imaginary part of permittivity. It is convenient to quantify this requirement by k/n ratio, where k is extinction coefficient and n is refractive index of the metal. High-quality plasmonic material possesses this k/n ratio high. This value usually decreases for ultrathin films due to scattering of free electrons on the layer interfaces and on the grain boundaries. It is demonstrated in Ref. [19] where evolution of the optical constants of Al during the layer growth is studied. We deposited a number of Al layers with the targeted thickness of 17 nm. Al is highly reactive metal; therefore, the Al coating quality reflects the concentration of residual gases during the deposition process. The optical constants of sputtered Al with different quality are shown in Fig. 4. Curve of sample A nearly coincides with the optical constants of evaporated Al [20] (dotted line) with exception of interband transition around 800 nm. The sample B exhibits rather lower extinction coefficient, k, and slightly enhanced refractive index with respect to evaporated Al. Consequently, the k/n ratio of the sputtered Al layers is 8.9 and 5.5 at 403 nm for samples A and B, respectively. For the evaporated aluminum [20], this k/n ratio reaches value 9.8 at this wavelength.
Fig. 4

Optical constants of sputtered Al layers. The dotted curve exhibits data of evaporated Al from [20]

High quality of the Al layer plays a key role for the electric field enhancement and the photoluminescence excitation. Increased losses of the metal layer cause damping of the surface plasmon resonance. This effect can be obviously seen in Fig. 5 where is shown angular dependence of reflectance of Al layer in Kretschmann configuration. The measured curves of the two different samples are shown. The curves are compared with idealized simulation calculated by Fresnel equation using optical constants of the evaporated Al [20]. The native oxide layer of 3 nm is considered in this simulation. The simulated spectra exhibit narrow SPR peak at incident angle of 45°. The better of the Al layers (labeled as sample A) possesses the resonant peak at the identical position as the simulation. It is seen that low-quality Al layers (sample B) exhibit shift of SPR peak to higher value of incident angle. Simultaneously, the width of the peak is broadened. The lowered reflectance of measured samples with respect to simulation is caused by surface roughness, which is not considered in the idealized simulation.
Fig. 5

Dependence of reflectance of the sputtered Al layers on incident angle. The reflectance was measured in Kretschmann configuration at wavelength of 403 nm. The dotted curve exhibits simulation based on optical constants of evaporated Al shown in Fig. 3

Initially, the LiF layers were deposited on silicon substrates at substrate temperature ranging from room temperature to 300 °C. The layer thickness was in the range from 160 to 850 nm. Figure 6 shows surface morphology of the LiF layers revealing their polycrystalline nature. The film deposited at room temperature exhibits fine crystalline pattern while there is obvious enhancement of the crystal size for the sample deposited at 300 °C. The good crystallinity of all deposited samples is confirmed by X-ray diffraction measurement showing mainly (111) and (200) crystal orientation. The crystallite size was estimated to be greater than 110 nm (without including the strain broadening effect). Optical constants were estimated using spectral ellipsometry. The refractive index of LiF layer is estimated 1.415 at wavelength of 400 nm that is a slightly higher value than that published in Ref. [20] (1.399). Composition analysis was accomplished on 40 nm thick LiF layer deposited on the Al-coated substrate. Thus, surface charging was partially minimized. Nevertheless, the obtained spectra were shifted by about 4 eV due to charging effect. There are F1s, Li1s, and F2s peaks seen in the spectra shown in Fig. 7. The spectra exhibit very little surface contamination by adsorbed oxygen and carbon (hydrocarbon) with estimated atomic concentrations of 0.3 % and 2.5 %, respectively. The composition was calculated using the Li1s and F2s peaks shown in inset graph in Fig. 7. The obtained atomic concentration was estimated 60 and 37 % for Li and F, respectively.
Fig. 6

AFM image of the LiF films deposited at a room temperature and b 300° C. The full color intensity scale represents height a 50 nm and b 80 nm

Fig. 7

XPS analysis of the LiF layer (40 nm) deposited on Al-coated fused silica substrate

For measurement of SPR-excited photoluminescence, we prepared series of samples with the variable LiF thickness ranging from 20 to 71 nm. The LiF layers were deposited on the Al-coated fused silica substrate with the Al layer thickness 17 nm. The measured angular-dependent reflectance is shown in Fig. 8a, b at the wavelength of 403 and 445 nm, respectively. The SPR is characterized by the reflectance dip, which is below 8 % for all the samples at the wavelength of 403 nm, but its position shifts to the higher incident angle from 45° to about 64° with the increase in thickness of the LiF layer from 0 nm (bare Al layer) to 71 nm. Simultaneously, the SPR exhibits broadening of the reflectance dip with the increase in the LiF thickness that is consistent with the simulation on Fig. 2. Position of the dips is similar for the wavelength of 445 nm. But, the dips are shallower with the reflectance reaching 25, 20, and 15 % for the layer thickness of 20, 38, and 71 nm, respectively. This effect can be due to strong absorption of CC absorption band.
Fig. 8

Measured reflectance of bare Al layer and LiF/Al structure in Kretschmann configuration for thickness of LiF layer 20, 38, and 71 nm. The incident laser beam wavelength is a 403 nm and b 445 nm

The same experimental setup was used for the photoluminescence measurement. The photoluminescence was measured at the reflectance dip where the SPR excitation is maximal. We indicated only very weak photoluminescence peaking around 500 nm for the excitation wavelength of 403 nm. This observation is consistent with the fact that the used wavelength is at the shoulder of the excitation band of F3 +, while the F2 excitation band is outside the used wavelength [4]. The photoluminescence is significantly enhanced after changing the excitation wavelength to 445 nm. The color map in Fig. 9 shows the dependence of the photoluminescence intensity on the angle of incidence and the photoluminescence wavelength. There is a residual light of the excitation laser at 445 nm seen in this image. The edge at 475 nm corresponds to the long-pass edge filter. The maximal photoluminescent signal is observed at the incident angle around 61° that corresponds to highest efficiency of the excitation at the maximal SPR in the reflectance dip.
Fig. 9

Measured dependence of photoluminescence on incident angle and photoluminescence wavelength. The photoluminescence of the sample LiF(58 nm)/Al(17 nm) was excited in surface plasmon resonance mode at 445 nm

Figure 10 shows photoluminescence measurement of the samples with thickness of 20, 38, and 71 nm using the excitation wavelength of 445 nm. Each sample is measured at the incident angle of maximal SPR excitation that was previously found by angle-resolved reflectance measurement shown in Fig. 8. In the photoluminescent spectra, there is depicted a dominant band at 655 nm and a minor band at 530 nm corresponding to the F2 and F3 + color centers, respectively. The positions of these two bands are slightly shifted to lower wavelength with respect to those reported, e.g., in [1, 4]. The highest photoluminescent signal is recorded for the LiF layer with the thickness of 38 nm, while the 71-nm-thick sample exhibits the signal lowered by about 30 %. There is noticeable difference in the F2 and F3 + bands ratio between the two samples. The photoluminescence of the 20 nm thick layer is very weak disappearing in the background signal. The SPR-excited photoluminescence is compared with the directly excited one in Fig. 10 (dotted curve). We prepared a 40-nm-thick LiF layer on fused silica substrate for that purpose. The arrangement of the experimental setup (especially, the position of the collector of the emitted light) was kept the same as for the SPR-excited luminescence measurement except the direction of the excitation laser beam, which was directed to the sample at the angle of 30° to the surface normal. It is seen that the directly excited photoluminescence is much lower with respect to the SPR-excited sample of similar thickness (38 nm).
Fig. 10

SPR-excited photoluminescence of LiF layers with different thickness: 20, 38, and 71 nm. It is compared with direct excitation of photoluminescence of LiF layer (40 nm) on fused silica substrate (dotted line). The photoluminescence was excited in surface plasmon resonance mode at 445 nm

5 Discussion

Comparing the results of the two samples with 38- and 71-nm-thick LiF layer, we can relate the reflectance measurement with the photoluminescence intensity. In accordance with the Ref. [10], the increase in the adjacent photoluminescent layer on the Al layer leads to the decrease in electric field enhancement in the Al layer vicinity. This is associated with broadening of the reflectance dip at the SPR condition as it can be observed from Fig. 7. This effect leads to the decrease in photoluminescence that is consistent with our observation. Accordingly, we can expect the highest photoluminescent signal for the 20-nm-thick LiF layer. In the contrary to our expectation, the photoluminescence of this sample is very weak. Obviously, the strong quenching effect leads to non-radiative decay due to energy transfer from excited luminophore to the metallic Al under plasmon excitation [21]. We can expect that this quenching occurs in the vicinity of the metallic layer of the thicker samples as well. Consequently, mainly the top layer and the layer surface are involved in the luminescence emission of those samples. With respect to the enhanced surface roughness (see Fig. 6), we can conclude that the luminescence emission originates mainly from CCs aggregated at the surface and the grain boundaries of the LiF layers. The surface CC of different alkali halides was already studied [22, 23]. Their optical characteristics differ from that of the bulk. Increase in the refractive index due to enhanced CC concentration was studied in [3]. The slightly enhanced refractive index of our samples is in accordance with that observation confirming the CC defects in our LiF layers.

Our experiments show that there is a significant photoluminescence of CCs revealing high concentration of the aggregated anion vacancy defects in “as deposited” LiF layers. The composition analysis confirms that there is lack of fluorine in the LiF layers. The missing fluorine atoms are prerequisite for the formation of anionic vacancies and their aggregation in the LiF structure. The strong photoluminescence of “as deposited” LiF layers can be explained by high concentration of CCs due to escape of volatile fluorine during the deposition process. In different publications, there are shown the similar photoluminescence spectrum with the dominant F2 and the minor F3 + CC after irradiation by gamma rays [4] or X-rays [24]. We did not observe an increase in the photoluminescence of our samples upon the X-ray irradiation showing that the concentration of CCs of as-deposited samples is saturated.

6 Conclusions

We proved that the surface plasmon resonance can be used for efficient excitation of the photoluminescence of very thin LiF layers. Aluminum is one of the most appropriate metals for this method, but its optical constants are critical for the efficient SPR coupling leading to high photoluminescence emission enhancement. The evaporated thin LiF layers exhibit polycrystalline nature. The high concentration of defects in non-irradiated LiF layers is revealed by intensive F2 and F3 + color centers’ emissions. Strong quenching was observed for the thinnest LiF layer due to non-radiative decay at vicinity of Al surface.

Notes

Acknowledgments

This research was supported by Czech Science Foundation, project GAP108/11/1312 and by Ministry of Education Youth and Sports, project LM2011029.

References

  1. 1.
    G. Baldacchini, J. Lumin. 100, 333 (2002)CrossRefGoogle Scholar
  2. 2.
    M. Montecchi, E. Nichelatti, A. Mancini, R.M. Montereali, J. Appl. Phys. 86, 3745 (1999)ADSCrossRefGoogle Scholar
  3. 3.
    E. Nichelatti, F. Bonfigli, A.Y. Faenov, F. Flora, T. Marolo, M. Montecchi, R.M. Montereali, T. Pikuz, G. Baldacchini, J. Non Cryst. Solids 351, 1774 (2005)ADSCrossRefGoogle Scholar
  4. 4.
    M.A. Vincenti, T. Marolo, R.M. Montereali, Radiat. Meas. 45, 359 (2010)CrossRefGoogle Scholar
  5. 5.
    R.M. Montereali, F. Bonfigli, M. Piccinini, E. Nichelatti, M.A. Vincenti, J. Lumin. 170, 761 (2016)CrossRefGoogle Scholar
  6. 6.
    J. Homola, S.S. Yee, G. Gauglitz, Sens. Actuators B 54, 3 (1999)CrossRefGoogle Scholar
  7. 7.
    M. Kikawada, A. Ono, W. Inami, Y. Kawata, Appl. Phys. Lett. 104, 223703 (2014)ADSCrossRefGoogle Scholar
  8. 8.
    E. Kretschmann, Z. Phys. 241, 313 (1971)ADSCrossRefGoogle Scholar
  9. 9.
    A. Otto, Z. Phys. 216, 398 (1968)ADSCrossRefGoogle Scholar
  10. 10.
    A. Ono, M. Kikawada, R. Akimoto, W. Inami, Y. Kawata, Opt. Express 21, 17447 (2013)ADSCrossRefGoogle Scholar
  11. 11.
    J. Enderlein, T. Ruckstuhl, Opt. Express 13, 8855 (2005)ADSCrossRefGoogle Scholar
  12. 12.
    N. Calander, J. Phys. Chem. B 109, 13957 (2005)CrossRefGoogle Scholar
  13. 13.
    S.D. Choudhury, R. Badugu, J.R. Lakowicz, Acc. Chem. Res. 48, 2171 (2015)CrossRefGoogle Scholar
  14. 14.
    I. Gryczynski, J. Malicka, K. Nowaczyk, Z. Gryczynski, J.R. Lakowicz, J. Phys. Chem. B 108, 12073 (2004)CrossRefGoogle Scholar
  15. 15.
    H.A. Macleod, Thin Film Optical Filters, 3rd edn. (Institute of Physics Publishing, Bristol, 2001)CrossRefGoogle Scholar
  16. 16.
    P.K. Maharana, S. Bharadwaj, R. Jha, J. Appl. Phys. 114, 014304 (2013)ADSCrossRefGoogle Scholar
  17. 17.
    H.R. Gwon, S.H. Lee, Mater. Trans. 51, 1150 (2010)CrossRefGoogle Scholar
  18. 18.
    F.-C. Chien, S.-J. Chen, Biosens. Bioelectron. 20, 633 (2004)CrossRefGoogle Scholar
  19. 19.
    M. Novotny, J. Bulir, J. Lancok, P. Pokorny, M. Bodnar, J. Nanophotonics 5, 051503 (2011)ADSCrossRefGoogle Scholar
  20. 20.
    E.D. Palik, Handbook of Optical Constants, vol. 1 (Academic Press, New York, 1985)Google Scholar
  21. 21.
    P. Anger, P. Bharadwaj, L. Novotny, Phys. Rev. Lett. 96, 113002 (2006)ADSCrossRefGoogle Scholar
  22. 22.
    V. Zielasek, T. Hildebrandt, M. Henzler, Phys. Rev. B 62, 2912 (2000)ADSCrossRefGoogle Scholar
  23. 23.
    A.P. Voitovich, V.S. Kalinov, A.P. Stupak, A.N. Novikov, L.P. Runets, J. Lumin. 157, 28 (2015)CrossRefGoogle Scholar
  24. 24.
    F. Bonfigli, M.A. Vincenti, S. Almaviva, R.M. Montereali, E. Nichelatti, R.N. Nogueira, H.J. Kalinowski, Appl. Opt. 48, G38 (2009)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2016

Authors and Affiliations

  1. 1.Institute of PhysicsAcademy of Sciences of the Czech RepublicPrague 8Czech Republic

Personalised recommendations