Open Access
Add to BibSonomyAbstract
Titanium oxynitride (TiNxOy) thin films are deposited by RF sputtering with various N2 flow rates. Two distinct absorption bands, one in the photon energies of ~0.5 – 2.5 eV and the other in the energies of ~3.5 – 5 eV, which correspond to the localized surface plasmon resonance (LSPR) of the TiN nanoparticles in the thin films and the interband transitions in TiNxOy, respectively, are observed in the absorption measurement. A Drude-Lorentz model, including the contributions of the free electrons, LSPR and interband transitions, is able to well fit the spectroscopic ellipsometric (SE) data. The resonance energy and strength of the LSPR oscillator are accurately determined from the SE analysis. The resonance energy is in the range of ~1 – 1.3 eV and blue-shifts with increasing N2 flow rate; and the strength decreases significantly with increasing N2 flow rate. The plasma energy yielded from the SE analysis shows a correlation with the conduction electron concentration obtained from the Hall effect measurement. It is shown that the LSPR plays a significant role in the complex dielectric function of the TiNxOy grains at the low photon energies (~0.5 – 1.5 eV).
© 2016 Optical Society of America
1. Introduction
Titaniun oxynitride (TiNxOy) thin films have been intensively studied due to the extraordinary optical and electrical properties, and chemical stability, etc [1,2]. The physical properties of TiNxOy thin films are between metallic TiN and insulated TiO2, and can be tuned by changing the ratio of N to O [3,4]. Due to their unique optical properties, TiN and TiNxOy thin films have been used as an absorber layer in selective solar absorber (SSA), which should absorb the solar radiation as much as possible and emit the infrared radiation as low as possible [5–7]. TiN and TiNxOy films are also exceptionally promising plasmonic materials with tunable optical properties [8–10]. It has been reported that the approach for depositing TiNxOy films by magnetron sputtering from a TiN target with only one reactive gas of O2 has some advantages, e.g., the N/O ratio can be continuously controlled, the deposition process is more reproducible and stable [7]. On the other hand, TiNxOy films with around 30% of oxygen content can be also formed by reactive sputtering of Ti target with Ar/N2 without purposeful introduction of O2 gas due to the existence of residual oxygen in the deposition chamber [11]. Such TiNxOy films can be considered as a composite film containing metallic component (TiN) and dielectric (TiO2); and the optical properties of the TiNxOy films could be dominated by the metallic TiN component.
When a large number of free electrons are confined in nano-structured materials (e.g., transition metal nitrides, metals, doped semiconductors, conductive metal oxides, etc.), localized surface plasmon resonance (LSPR) could be excited by optical radiation, leading to strong light absorption and scattering, and the LSPR depends on the size and shape of the structures and the dielectric environment [10, 12–15]. It has been reported that the TiN nanodisks with the diameters of 60 – 180 nm and disk thickness of 30 nm formed by electron-beam lithography show the LSPR with the resonance wavelength in the range of ~800 – 1200 nm depending on the diameter of the nanodisks [16]. Semi-shell TiN nano-structures exhibit a tuneable localized plasmon resonance with light [17]. Due to the plasmonic resonance in visible light range and intrinsic loss, a TiN absorber achieves an average absorption of 95% in the wavelength range of 400 – 800 nm [18].
In TiNxOy films, the size of the TiN nanoparticles affects the electrical and electronic properties of the films, as it determines the mean free path of conduction electrons (electron scattering at the grain boundaries) and the free-electron concentration [19]. With a high concentration of free electrons in the TiN nanoparticles, LSPR could be also excited with light at the resonance energy. On the other hand, as an intermetallic compound, the optical properties of TiN are very sensitive to the stoichiometry [19,20]. In the present study, TiN nanoparticles are formed in the TiNxOy thin films deposited by RF sputtering of TiN target with zero oxygen gas flow rate, and both the nitrogen and oxygen contents in the films are changed by varying the N2 flow rate during the deposition process, which leads to the change in the free-electron concentration. Absorbance measurement on the TiNxOy thin films shows that there is an absorption band in the photon energy range of ~0.5 – 2.5 eV, which is attributed to the LSPR in the TiN nanoparticles. Information of the LSPR behaviours has been obtained from the spectroscopic ellipsometric study, and the influence of the LSPR on the complex dielectric function of the TiNxOy has been examined also.
2. Experiment
TiNxOy thin films with the thickness of ~60 nm were deposited on the following substrates for different types of measurements: a p-type Si substrate for the measurements of X-ray photoelectron spectroscopy (XPS), Hall effect, atomic force microscopy (AFM) and ellipsometry; a double-side polished silica substrate for the absorbance measurement; and a double side polished sodium chloride substrate for the transmission electron microscopy (TEM) characterization. The TiNxOy thin film deposition was carried out by RF magnetron sputtering of a pure 2 inch TiN target (>99.9% purity) with a Denton desktop sputtering system at room temperature in Ar/N2 environment. The deposition chamber was pumped down to 9 × 10−6 Torr. During the deposition, the Ar flow rate was maintained at 10 sccm, while the N2 flow rate was set at 0, 3, 6 and 9 sccm, respectively. The sputtering power was maintained at 100 W.
The chemical compositions of the TiNxOy thin films were examined with XPS (ESCALAB 250). Before the XPS experiment, the sample surface was cleaned with 4 keV Ar ion beam for 90 seconds. The XPS spectra were calibrated by using the C 1s peak (285 eV) as the reference. Figure 1 shows the XPS analysis for the TiNxOy thin film deposited only with Ar flow rate of 10 sccm. It can be concluded from the XPS analysis that the Ti-O, Ti-N-O, and Ti-N states exist in the TiNxOy thin films. Although the films are deposited by sputtering a pure TiN target only with Ar/N2 gases, oxygen still presents in the films, which is due to the existence of residual oxygen in the deposition chamber [11, 21]. Oxygen has stronger reaction ability than that of nitrogen with Ti and TiN [7]; and this oxygen would react with titanium in the films to make oxides. Therefore, the TiNxOy films of this work can be considered as a TiN/oxides composite. The relative atomic percentages of Ti, N, and O in the TiNxOy thin films obtained from the XPS analysis are presented in Table 1. The presence of large amounts of oxygen content in the TiNxOy thin films deposited by a sputtering process without purposeful introduction of O2 gas was also reported in other works [11, 22]. With the increase of N2 flow rate from 0 to 9 sccm, the ratio of N to O in TiNxOy thin film decreases from 0.52 to 0.19, indicating that TiN content in the films decreases with increasing N2 flow rate. This is attributed to the decrease of deposition rate with increase of N2 flow rate as a result of the nitriding effect of the TiN target by N2 [11, 23]. With the decrease of deposition rate (in the present work, the deposition rate decreases from 2.29 to 1.14 nm/min with the increase of N2 flow rate from 0 to 9 sccm.), the residual oxygen has more time to react with the Ti atoms and thus more oxygen is presented in the deposited films.
Fig. 1 XPS measurement of the TiNxOy thin film deposited only with Ar flow rate of 10 sccm: (a) survey scan, (b) Ti 2p, (c) N 1s, and (d) O 1s.
Table 1. Relative atomic percentages of Ti, N, and O in the TiNxOy thin films obtained from the XPS analysis
The concentration of conduction electrons of the TiNxOy thin films was measured with a Hall effect measurement system (HL5500). The concentration of conduction electrons as a function of N2 flow rate is shown in Fig. 2. When the N2 flow rate is increased from 0 to 9 sccm, the concentration decreases from 1.31 × 1019 cm−3 to 1.19 × 1018 cm−3. The conduction electrons of TiNxOy are mostly contributed by the TiN component. As shown in Table 1, the ratio of N to O of the TiNxOy thin films decreases with increasing N2 flow rate, indicating that the TiN content decreases with increasing N2 flow rate. This explains the decrease of the conduction-electron concentration with increasing N2 flow rate.
Fig. 2 Conduction-electron concentrations of the TiNxOy thin films obtained from the Hall effect measurement.
The surface morphology of the as-deposited TiNxOy thin films was characterized with AFM (Bruker, Dimension Icon) in the tapping mode using a Si probe. The AFM images of the thin films on the Si substrate deposited with various N2 flow rates are shown in the insets of Fig. 3. The AFM images show the existence of TiNxOy grains with the average size in the range of 14 – 25 nm in the thin films. The structural properties of the TiNxOy thin films are also characterized with high resolution TEM (HRTEM) (JEOL 2010) and selected area diffraction (SAD). As shown in Fig. 3(f), the HRTEM image indicates the existence of nanocrystalline structures in the TiNxOy thin film. The SAD pattern of the nanocrystalline structures is shown in the inset of Fig. 3(f). The TiN crystal orientations including (111), (200) and (220) can be identified in the SAD pattern (JCPDS 38-1420), suggesting that the nanocrystalline structures are TiN nanocrystals.
Fig. 3 Experimental absorbance spectra in the wavelength range of 250 – 2500 nm for the TiNxOy thin films deposited with various N2 flow rates: (a) 0 sccm, (b) 3 sccm, (c) 6 sccm, and (d) 9 sccm. The inserts in Figs. 3(a)–3(d) show the AFM images of the TiNxOy thin films. (e) Normalized absorbance spectra of the TiNxOy thin films in the photon energy range of 0.5 – 5 eV. The insert in Fig. 3(e) is a schematic illustration of electron cloud oscillating in the TiN nanoparticles under the influence of changing electric field of light. (f) HRTEM image of the TiNxOy thin films deposited only with Ar. The insert in Fig. 3(f) is the corresponding SAD pattern.
Absorbance spectra of the TiNxOy thin films deposited on silica substrate were measured with an UV-Vis-NIR spectrophotometer (PerkinElmer Lambda 950) in the wavelength range of 250 – 2500 nm. The absorbance spectra for the N2 flow rates of 0, 3, 6, 9 sccm are shown in Figs. 3(a)–3(d), respectively, and a comparison of the spectra is shown in Fig. 3(e). As can be observed in Fig. 3, in the measured photon energy range of 0.5 – 5 eV (wavelength range of 250 – 2500 nm), all the absorbance spectra exhibit two distinct absorption bands, i.e., one in the low photon energy range of ~0.5 – 2.5 eV (wavelength range of ~500 – 2500 nm) and the other in the high photon energy range of ~3.5 – 5 eV (wavelength range of ~250 – 350 nm). With the increase of the N2 flow rate, the absorption band in the low photon energy range becomes weaker and shows a blue shift; while the absorption band in the high photon energy range does not show very significant change. As discussed later, the absorption band in the low photon energy range is attributed to the LSPR of the TiN nanoparticles in the TiNxOy thin films while the large absorbance in the high photon energy range is due to the interband transitions in the TiNxOy thin films.
Spectroscopic ellipsometry (SE) is a powerful technique with very high precision for optical-property characterization [24]. The SE measurements on the TiNxOy thin films deposited on Si substrate with various N2 flow rates were conducted with an ellipsometer (J.A. Woollam VB-250) in the wavelength range of 350 – 1500 nm at the angle of incidence of 75°, and the ellipsometric spectra are shown in Fig. 4. An ellipsometric analysis was carried out based on an optical dispersion model taking the contributions of the conduction electrons, LSPR in the TiN nanoparticles and interband transitions in the TiNxOy into account. The result yielded from the ellipsometric analysis is consistent with that of the absorbance study, and the ellipsometric analysis also highlights the effect of the conduction electrons in the TiN nanoparticles on the LSPR.
Fig. 4 Ellipsometric fittings to the experimental ellipsometric angles (Ψ and Δ) for various N2 flow rates. The insert in Fig. 4(a) shows the three-phase model (i.e., air/TiNxOy layer (grains + air voids)/Si substrate) used in the ellipsometric modeling.
3. Results and discussion
The Hall effect measurement shows that there is a high concentration of free electrons in the TiN nanoparticles inside the TiNxOy grains. With such high concentration of free electrons in the TiN nanoparticles (the mean free path of conduction electrons is determined by the nanoparticle size), LSPR can be excited with light at the resonance energy, as illustrated in the insert of Fig. 3(e). It has been reported that for 30 nm thick 400 ̊C-grown TiN nanodisk arrays formed by electron-beam lithography, the LSPR resonance wavelength is in the range of ~800 – 1200 nm depending on the diameter of the nanodisks (60 – 180 nm) [16]. Therefore, the absorption band in the low photon energy range shown in Fig. 3 is attributed to LSPR in the TiN nanoparticles with the resonance wavelength in the range of ~960 −1100 nm (photon energy of ~1.12 −1.29 eV) depending on the N2 flow rate. As expected, it is observed that a higher concentration of free electrons leads to a higher LSPR peak intensity. This is consistent with the observation that LSPR in quantum dots of semiconductor copper (i) sulphide is largely affected by the free carriers in the quantum dots, i.e., the LSPR absorbance intensity increases with the free-carrier concentration [14].
A Drude-Lorentz model is usually used to describe the optical dispersion of TiN, where the Drude term and the Lorentz oscillators represent the intraband and interband transitions, respectively [19, 25]. There are various possible interband transitions in TiN, such as the transitions at the Г15 - Г12, Х5 - Х2, L3 - L3’ points of the Brillouin zone which correspond to strong interband absorption at ~2.3, 3.9, 5.6 eV, respectively [19]. For TiNxOy thin film, the interband transitions may be dependent of the chemical composition and the nanocrystalline structures of the thin film. In this work, the large absorbance in the high photon energy range (3.5 – 5 eV) shown in Fig. 3(e) can be attributed to the interband transitions in TiNxOy. Here one Lorentz oscillator is used to represent the interband transitions. As discussed below, the modeling involving the Lorentz oscillator agrees well with the experimental results of ellipsometric spectra in the high photon energy range where the interband transitions occur. This indicates that one Lorentz oscillator can adequately represent the interband transitions in the photon energy of 3.5 – 5 eV. As the TiNxOy thin films exhibit a high concentration of conduction electrons, the Drude term is used to characterize the optical response of the free electrons in the material. Taking the contributions of the free electrons, LSPR in the TiN nanoparticles and interband transitions in the TiNxOy into account, the complex dielectric function (ε = εr + iεm, where εr and εm are the real and imaginary parts of the complex dielectric function ε, respectively) of the TiNxOy thin film is given by
where ε∞ is the high frequency dielectric constant. The contribution of the free electrons is described by the Drude termwhere Ep is the plasma energy; E is the photon energy; ГD is the Drude damping factor. The contributions of the LSPR and interband transitions are represented by the following two Lorentz oscillators, respectively, where fLSPR, ELSPR and ГLSPR are the strength, resonance energy and damping (broadening) factor of the oscillator for the LSPR, respectively; and fi, Ei and Гi are the strength, resonance energy and damping (broadening) factor of the oscillator for the interband transitions, respectively. The real part (εr) and imaginary part (εm) of the complex dielectric function can be calculated with Eqs. (1)–(4).An ellipsometric modeling is carried out to fit the ellipsometric data (ellipsometric angles Ψ and Δ) shown in Fig. 4. In the ellipsometric modeling, the TiNxOy thin film is treated as an effective medium layer consisting of TiNxOy grains and air voids as there may be some nano-gaps between the grains in the TiNxOy thin film; and the three-phase model (i.e., air/TiNxOy grains with air voids/Si substrate) is used, as illustrated in the insert of Fig. 4(a). The effective complex dielectric function of the TiNxOy layer can be calculated with the Bruggeman effective medium approximation (B-EMA) [26]
where εi is the effective complex dielectric function of the TiNxOy layer (i.e., TiNxOy grains + air voids); εTiNxOy is the complex dielectric function of the TiNxOy grains; εair is the dielectric function of air; f is the volume fraction of TiNxOy grains in the TiNxOy thin film. The complex dielectric function εTiNxOy of the TiNxOy grains is calculated with Eqs. (1)–(4), with the contributions of the intraband transition (the conduction electrons), interband transitions (the bound electrons) and the LSPR due to the surface-bound charge density oscillations of the TiN nanoparticles taken into account. The model parameters for the Drude term and Lorentz oscillators for the LSPR and interband transitions, layer thickness and air volume fraction are the fitting parameters in the fitting to the experimental ellipsometric angles (Ψ and Δ) in a wide wavelength range (in this work, 350 – 1500 nm). A procedure for ellipsometric fitting can be found in other studies [26, 27]. The ellipsometric fittings for the N2 flow rates of 0, 3, 6 and 9 sccm are shown in Figs. 4(a)–4(d), respectively. Excellent agreement between the modeling and experiment has been achieved for all the N2 flow rates.The LSPR resonance energies (ELSPR) obtained from the ellipsometric fittings for various N2 flow rates are shown in Fig. 5. As can be observed in the figure, ELSPR increases with the N2 flow rate, which could be due to the changes in the structural properties and dielectric environment (which could be affected by the non-conductive composition, e.g., TiO2 in the TiNxOy thin films) of the TiN nanoparticles. It is well known that LSPR strongly depends on particle size, shape, and inter-particle spacing, which directly influence the restoring forces of oscillating electrons inside the particle [28,29]. Numerical simulations show a strong dependence of the resonant energy on the morphology of noble metal particles [30,31]. The size effect on the LSPR resonance wavelength was experimentally demonstrated in the study reported in [16], where the well-defined TiN nanodisk arrays were formed by electron-beam lithography. Blue-shift of the LSPR peak with decreasing particle size observed in the study, e.g., for the TiN nanodisks grown at 400 ̊C, the LSPR peak absorption occurs at the wavelengths of ~800 nm (~1.55 eV) and ~1200 nm (~1.03 eV) for the smallest particles (60 nm) and the largest particles (200 nm), respectively [16]. In the present study, the average grain size as obtained from the AFM measurement decreases from ~25 nm to ~14 nm as the N2 flow rate increases from 0 to 9 sccm. Although it is difficult to measure the actual TiN particle size, the grain size measurement suggests that the size of the particles responsible for the LSPR decreases with increasing N2 flow rate, which is also supported by the results shown in Table 1 that TiN content in the films decreases with increasing N2 flow rate. Correspondingly, the LSPR resonance energy increases from 1.03 eV to 1.27 eV as the N2 flow rate increases from 0 to 9 sccm. This is consistent with the general observation of blue-shift of plasmon for decreasing size.
The strength of the LSPR oscillator (fLSPR) obtained from the ellipsometric fittings for various N2 flow rates are shown in Fig. 6. The dependence of the normalized absorbance peak intensity on the N2 flow rate is also shown in the figure. It can be observed from the figure that the strength of the LSPR oscillator decreases with N2 flow rate, being consistent with the dependence of the normalized absorbance peak intensity on N2 flow rate. This actually reflects the fact that the free-electron concentration of the TiNxOy thin films decreases with N2 flow rate. As the concentration of the free electrons in the TiNxOy is lower, the LSPR absorbance intensity is lower and the strength of the LSPR oscillator is smaller.
Fig. 6 LSPR strength (fLSPR) and the normalized absorbance peak intensity as a function of N2 flow rate.
On the other hand, another parameter used in the ellipsometric modeling that is directly related to the free-electron concentration is the plasma energy Ep. The relationship between Ep and the free electron concentration Nv is given by
where q is the electronic charge; ε0 is the free space permittivity; and m* is the electron effective mass. The values of Ep obtained from the ellipsometric fittings for the various N2 flow rates are shown in Fig. 7(a). For comparison, the conduction-electron concentration obtained from the Hall effect measurement is also shown in the figure. As shown in Fig. 7(a), both Ep and the conduction-electron concentration decrease with N2 flow rate. Thus the decrease of Ep with N2 flow rate can be attributed to the fact that there are less free electrons in the TiNxOy thin film deposited with a higher N2 flow rate. Assuming the following simple relationship between the free electron concentration and the conduction-electron concentration (Nc), Nv = N0 + Nc where N0 is the concentration of those free electrons that do not contribute in the Hall effect measurement, one can expect the linear relationship between Ep2 and Nc expressed bywhere α = q2/(ε0m*). The above assumption is reasonable as explained in the following. The free electrons in the metallic nanoparticles respond to the optical excitation; however, not all of the free electrons contribute in the Hall effect measurement because some of the free electrons are confined in the nanoparticles (note that the metallic nanoparticles are embedded in the dielectric) and are not able to contribute in the current conduction. Therefore, the conduction-electron concentration (Nc) yielded from the Hall effect measurement should be smaller than the free electron concentration that is determined from the plasma energy Ep. The linear relationship between Ep2 and Nc is observed in Fig. 7(b). The value of N0 obtained from the linear fittings is 2.93 × 1018cm−3.Fig. 7 (a) Ep as a function of N2 flow rate. The conduction-electron concentrations (Nc) obtained from the Hall effect measurement are also included for comparison. (b) Ep2 versus Nc.
The LSPR of the free electrons in the TiN nanoparticles has a significant influence on the complex dielectric function of the TiNxOy grains in the low photon energy range of 0.5 – 2.5 eV. Figure 8(a) shows the real part of the complex dielectric function due to the contribution of LSPR for various N2 flow rates. As can be observed in the figure, the real part of the complex dielectric function increases when photon energy decreases in the photon energy range of 0.5 – 1.5 eV (wavelength range of ~800 – 2500 nm); and such LSPR effect is reduced for a higher N2 flow rate as a result of the decrease in the free electron concentration. For bulk TiN in the photon energy range, the experimental real part of complex dielectric function is negative [25]. However, as shown in Fig. 8(b), the real part of the complex dielectric function of the TiNxOy grains deposited with 0 sccm N2 flow rate obtained from the above ellipsometric study is positive, which is partially due to the contribution of LSPR of the TiN nanparticles (the oxide composition may play an important role also), as demonstrated in Fig. 8(b). On the other hand, as shown in Fig. 9(a), the imaginary part of the complex dielectric function due to the LSPR contribution decreases with N2 flow rate, which is a result of the decrease in the free electron concentration, being consistent with the effect of N2 flow rate on the LSPR absorbance (Fig. 3(e)). Figure 9(b) shows that the imaginary part of the complex dielectric function of the TiNxOy grains in the low energy range (0.5 – 1.5 eV) is indeed significantly affected by the LSPR contribution.
Fig. 8 Real part of complex dielectric function in the low energy range of 0.5 – 2.5 eV: (a) the contribution of LSPR for various N2 flow rates; and (b) individual contributions of the LSPR and Drude term for 0 sccm N2 flow rate. The εr of the TiNxOy grains deposited with 0 sccm N2 flow rate and εr of bulk TiN are included in Fig. 8(b) for comparison.
Fig. 9 Imaginary part of complex dielectric function in the low energy range of 0.5 – 2.5 eV: (a) the contribution of LSPR for various N2 flow rates; and (b) individual contributions of the LSPR and Drude term for 0 sccm N2 flow rate. The εm of the TiNxOy grains deposited with 0 sccm N2 flow rate and εm of bulk TiN are included in Fig. 9(b) for comparison.
4. Conclusion
In conclusion, TiNxOy thin films have been deposited by RF-magnetron sputtering of TiN target with various N2 flow rates. The XPS characterization confirms that the Ti-N, Ti-O, and Ti-N-O chemical states presents in the TiNxOy thin films; the Hall effect measurement shows that the conduction electron concentration of the thin films decreases with the increasing of N2 flow rate; the AFM measurement suggests that TiNxOy thin films are formed by TiNxOy grains; and the TEM characterization shows the existence of TiN nanocrystals in the TiNxOy thin films. The absorbance spectra measured in the photon energy range of 0.5 – 5 eV (wavelength range of 250 – 2500 nm), exhibit two distinct absorption bands, i.e., one in the low photon energy range of ~0.5 – 2.5 eV (wavelength range of ~500 – 2500 nm) and the other in the high photon energy range of ~3.5 – 5 eV (wavelength range of 250 – 350 nm). With the increase of the N2 flow rate, the absorption band in the low photon energy range becomes weaker and shows a blue shift; while the absorption band in the high photon energy range does not show very significant change. The absorption band in the low photon energy range is attributed to the LSPR of the TiN nanoparticles in the TiNxOy grains; while the large absorbance in the high photon energy range is due to the interband transitions in TiNxOy. A model taking the contributions of the free electrons, LSPR and interband transitions into account, which are represented by a Drude term and two Lorentz oscillators, respectively, is able to well fit the ellipsometric spectra in the photon energy range of ~0.8 – 3.5 eV for various N2 flow rates. The resonance energy and strength of the LSPR oscillator are accurately determined from the ellipsometric modeling. The resonance energy is in the range of ~1 – 1.3 eV and blue-shifts with increasing N2 flow rate; and the strength decreases significantly with increasing N2 flow rate. The plasma energy yielded from the ellipsometric modeling shows a correlation with the conduction electron concentration obtained from the Hall effect measurement. It is shown that the LSPR plays a significant role in the complex dielectric function of the TiNxOy grains in the photon energy of 0.5 – 1.5 eV (wavelength range of ~800 – 2500 nm).
Funding
National Research Foundation (NRF-CRP13-2014-02); NTU-A*STAR Silicon Technologies Centre of Excellence (Program Grant No. 112 3510 0003); National Natural Science Foundation of China (61274086).
References and links
1. M. H. Chan and F. H. Lu, “Preparation of titanium oxynitride thin films by reactive sputtering using air/Ar mixtures,” Surf. Coat. Tech. 203(5–7), 614–618 (2008). [CrossRef]
2. J. Graciani, S. Hamad, and J. F. Sanz, “Changing the physical and chemical properties of titanium oxynitrides TiN1-xOx by changing the composition,” Phys. Rev. B 80(18), 184112 (2009). [CrossRef]
3. S. J. Cho, C. K. Jung, and J. H. Boo, “A study on the characteristics of TiOxNy thin films with various nitrogen flow rate by PECVD method,” Curr. Appl. Phys. 12, S29–S34 (2012). [CrossRef]
4. J. M. Chappé, N. Martin, J. Lintymer, F. Sthal, G. Terwagne, and J. Takadoum, “Titanium oxynitride thin films sputter deposited by the reactive gas pulsing process,” Appl. Surf. Sci. 253(12), 5312–5316 (2007). [CrossRef]
5. J. C. Francois and M. Sigrist, “The optical properties of titanium nitrides and carbides: spectral selectivity and photothermal conversion of solar energy,” Sol. Energy Mater. 7(3), 299–312 (1982). [CrossRef]
6. M. Lazarov, B. Röhle, T. Eisenhammer, and R. Sizmann, “TiNxOy-Cu-coatings for low emissive solar selective absorbers,” Proceedings of the Soc. Photo. Opt. Instrumentat. Engineers (SPIE) 1536, 183–191 (1991).
7. F. Chen, S. W. Wang, L. Yu, X. Chen, and W. Lu, “Control of optical properties of TiNxOy films and application for high performance solar selective absorbing coatings,” Opt. Mater. Express 4(9), 1833 (2014). [CrossRef]
8. G. V. Naik, J. Kim, and A. Boltasseva, “Oxides and nitrides as alternative plasmonic materials in the optical range,” Opt. Mater. Express 1(6), 1090–1099 (2011). [CrossRef]
9. G. V. Naik, J. L. Schroeder, X. J. Ni, A. V. Kildishev, T. D. Sands, and A. Boltasseva, “Titanium nitride as a plasmonic material for visible and near infrared wavelengths,” Opt. Mater. Express 2(4), 478–489 (2012). [CrossRef]
10. U. Guler, G. V. Naik, A. Boltasseva, V. M. Shalaev, and A. V. Kildishev, “Performance analysis of nitride alternative plasmonic materials for localized surface plasmon applications,” Appl. Phys. B-Lasers Opt. 107(2), 285–291 (2012). [CrossRef]
11. N. White, A. L. Campbell, J. T. Grant, R. Pachter, K. Eyink, R. Jakubiak, G. Martinez, and C. V. Ramana, “Surface/interface analysis and optical properties of RF sputter-deposited nanocrystalline titanium nitride thin films,” Appl. Surf. Sci. 292, 74–85 (2014). [CrossRef]
12. E. Hutter and J. H. Fendler, “Exploitation of localized surface plasmon resonance,” Adv. Mater. 16(19), 1685–1706 (2004). [CrossRef]
13. S. Zhu, T. P. Chen, Y. C. Liu, Y. Liu, and S. Fung, “A quantitative modeling of the contributions of localized surface plasmon resonance and interband transitions to absorbance of gold nanoparticles,” J. Nanopart. Res. 14(5), 856 (2012). [CrossRef]
14. J. M. Luther, P. K. Jain, T. Ewers, and A. P. Alivisatos, “Localized surface plasmon resonances arising from free carriers in doped quantum dots,” Nat. Mater. 10(5), 361–366 (2011). [CrossRef] [PubMed]
15. J. Kim, G. V. Naik, N. K. Emani, U. Guler, and A. Boltasseva, “Plasmonic resonances in nanostructured transparent conducting oxide films,” IEEE J. Sel. Top. Quant. 19(3), 1–7 (2013). [CrossRef]
16. U. Guler, J. C. Ndukaife, G. V. Naik, A. G. A. Nnanna, A. V. Kildishev, V. M. Shalaev, and A. Boltasseva, “Local heating with lithographically fabricated plasmonic titanium nitride nanoparticles,” Nano Lett. 13(12), 6078–6083 (2013). [CrossRef] [PubMed]
17. M. B. Cortie, J. Giddings, and A. Dowd, “Optical properties and plasmon resonances of titanium nitride nanostructures,” Nanotechnology 21(11), 115201 (2010). [CrossRef] [PubMed]
18. W. Li, U. Guler, N. Kinsey, G. V. Naik, A. Boltasseva, J. Guan, V. M. Shalaev, and A. V. Kildishev, “Refractory plasmonics with titanium nitride: broadband metamaterial absorber,” Adv. Mater. 26(47), 7959–7965 (2014). [CrossRef] [PubMed]
19. P. Patsalas and S. Logothetidis, “Optical, electronic, and transport properties of nanocrystalline titanium nitride thin films,” J. Appl. Phys. 90(9), 4725–4734 (2001). [CrossRef]
20. J. H. Kang and K. J. Kim, “Structural, optical, and electronic properties of cubic TiNx compounds,” J. Appl. Phys. 86(1), 346–350 (1999). [CrossRef]
21. N. K. Ponon, D. J. R. Appleby, E. Arac, P. J. King, S. Ganti, K. S. K. Kwa, and A. O’Neill, “Effect of deposition conditions and post deposition anneal on reactively sputtered titanium nitride thin films,” Thin Solid Films 578, 31–37 (2015). [CrossRef]
22. C. M. Zgrabik and E. L. Hu, “Optimization of sputtered titanium nitride as a tunable metal for plasmonic applications,” Opt. Mater. Express 5(12), 478–489 (2015). [CrossRef]
23. S. Wrehde, M. Quaas, R. Bogdanowicz, H. Steffen, H. Wulff, and R. Hippler, “Optical and chemical characterization of thin TiNx films deposited by DC-magnetron sputtering,” Vacuum 82(10), 1115–1119 (2008). [CrossRef]
24. H. Fujiwara, Spectroscopic Ellipsometry: Principles and Applications (Wiley, 2007), pp. 81–87. [CrossRef]
25. S. T. Sundari, R. Ramaseshan, F. Jose, S. Dash, and A. K. Tyagi, “Investigation of temperature dependent dielectric constant of a sputtered TiN thin film by spectroscopic ellipsometry,” J. Appl. Phys. 115(3), 033516 (2014). [CrossRef]
26. X. D. Li, T. P. Chen, P. Liu, Y. Liu, Z. Liu, and K. C. Leong, “A study on the evolution of dielectric function of ZnO thin films with decreasing film thickness,” J. Appl. Phys. 115(10), 103512 (2014). [CrossRef]
27. T. P. Chen, Y. Liu, M. S. Tse, P. F. Ho, G. Dong, and S. Fung, “Depth profiling of Si nanocrystals in Si-implanted SiO2 films by spectroscopic ellipsometry,” Appl. Phys. Lett. 81(25), 4724 (2002). [CrossRef]
28. X. D. Li, T. P. Chen, Y. Liu, and K. C. Leong, “Influence of localized surface plasmon resonance and free electrons on the optical properties of ultrathin Au films: a study of the aggregation effect,” Opt. Express 22(5), 5124–5132 (2014). [CrossRef] [PubMed]
29. S. Link and M. A. El-Sayed, “Optical properties and ultrafast dynamics of metallic nanocrystals,” Annu. Rev. Phys. Chem. 54(1), 331–366 (2003). [CrossRef] [PubMed]
30. H. Kuwata, H. Tamaru, K. Esumi, and K. Miyano, “Resonant light scattering from metal nanoparticles: Practical analysis beyond rayleigh approximation,” Appl. Phys. Lett. 83(22), 4625–4627 (2003). [CrossRef]
31. S. W. Prescott and P. Mulvaney, “Gold nanorod extinction spectra,” J. Appl. Phys. 99(12), 123504 (2006). [CrossRef]
