Open Access
Add to BibSonomy
Abstract
Room temperature, broadband infrared (IR) bolometer was investigated for the first time by using Mn1.56Co0.96Ni0.48O4 (MCNO) thin film with dielectric-metal-dielectric absorptive layers. The Si3N4/NiCr/SiO2 layer was constructed to improve light absorption. A responsivity of 98.6 V/W and D* of 2.1 × 107 cm∙Hz0.5/W@20 Hz, with a typical time constant of 14.5 ms, was obtained with a 1550 nm laser. The response spectra of the detector covered the range from near to far infrared, which greatly enhanced the potential of MCNO films in large-scale IR thermal detection applications. This study provides an efficient way to develop large scale, broadband MCNO IR detectors.
© 2019 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
Infrared thermal detectors and bolometers, which typically relies on electrical resistance variation caused by photon-thermal transition process, attract significant research interests in recent years, primarily due to their practical applications in many fields, including night vision, bio sensing, gas sensing, and so on [1–3].
Two major objectives in thermal sensitive bolometric detector design can be summarized as: (1) high absorption in the infrared (IR) range and thus improved sensitivity, without excessive demands in other properties such as the response speed; (2) broad working temperature range, in order to improve the adaptability of the device. Metal oxide materials serve as important candidates for thermal sensitive bolometric detectors. In recent years, IR bolometric detectors were fabricated using metal oxide materials such as VOx [4,5], La1-xAxMnO3 (A = Ca, Sr, Ba, Pb) [6,7], and Mn-Co-Ni-O spinels [8]. Among these materials, Mn-Co-Ni-O (which has a structure of AB2O4) shows many advantages over other materials because of its high value of negative temperature coefficient (NTC) (−3% to −5%K−1) in a broad temperature range of 230 K to 400 K [9], long term stability [10], and suitable resistivity (~102-103 Ω∙cm) to fabricate sensitive elements suitable for CMOS readout circuits [1]. Among different metallic ratios, Mn1.56Co0.96Ni0.48O4 (MCNO) is very near the resistivity minimum for the ternary oxide, as well as exhibits a strong absorption band in the far infrared range of 14-25 μm [11], which have been extensively investigated and widely used in earth radiation budget measurement, thermometry and uncooled infrared detection [12]. In recent years, MCNO thin films have been fabricated by a variety of techniques, which promotes their applications in thermal sensing and IR bolometric detection [13–15].
However, the absorption of the MCNO film is weak in IR atmospheric window range (3-5 μm and 8-12 μm), where the extinction coefficient k of MCNO film is of the order of about 0.01 in mid-infrared [16,17]. Thus, the MCNO detector will only cover a limited part of the concerned IR range. Therefore, additional absorptive layers should be added to improve the absorption of the MCNO detector. In previous reports, gold black absorption layer or organic black painting were frequently used, which help improving the response of metal oxide microbolometers [18]. On the other hand, absorption layers by using film coating techniques were also important choice in fabricating large scale bolometric detectors [19].
In this study, we designed a self-heating compensated, broad band, infrared bolometric detector based on MCNO films by using a dielectric-metal-dielectric absorptive layer. Sandwich structured Si3N4/NiCr/SiO2 was constructed to improve the absorption of MCNO thin film in the mid-wave infrared range. The response spectra covered a broad wavelength range between 1.33 to 25 μm. Moreover, the responsivity and D* of the bolometric detector was also greatly improved at the concerned wavelength range, with merely 10-20% increase in time constant.
2. Experimental details
MCNO films were prepared on the sapphire substrate by using a radio-frequency (RF) deposition method. The detailed preparation procedures were reported elsewhere [20]. The substrate temperature for the growth was set to be 750 ◦C, and the deposition duration was set to be 140 hours to obtain desired film thickness of 10 micrometers. The phase identification was studied by x-ray diffraction (XRD) in the (θ, 2θ) configuration using a RigaKu D/MAX-2550 x-ray diffractometer with Cu Kα radiation (λ = 1.5418 Å). The cross-sectional image of the film was identified by using the scanning electron microscopy (SEM). The transmission (T) and reflection (R) spectra of MCNO samples grown on sapphire substrate were measured by using a Fourier spectrometer (Bruker 80 V, Germany) in a range of 400 cm−1 to 7500 cm−1. The absorptance (Abs) spectra of the MCNO films grown on sapphire substrate with or without the absorption layers were studied and compared in order to evaluate the performance of the MCNO detector with absorptive layers.
Sandwich structured absorption layers of Si3N4/NiCr/SiO2 were deposited onto the MCNO thin films to improve the optical absorption in the mid-infrared absorption range. A silicon dioxide (SiO2) film was deposited onto the MCNO films by radio frequency (RF) magnetron sputtering with a thickness of 100 nm at 200 ◦C. Then, the NiCr absorptive layer was deposited by using a dual ion beam sputtering system (LDJ-2A-150F, Beijing). A metal alloy target of NiCr with metallic ratio of Ni:Cr = 80:20 was used, where a NiCr thin film with a thickness of about 15 nm was deposited as the absorption layer. Prior to film deposition, the base pressure was evacuated to be approximately 5 × 10−3 Pa, and the working pressure was 0.02 Pa. High purity (>99.99%) argon was applied to provide the plasma. The ion energy was set to be 500 eV and ion beam current was set to be 70 mA, where the sputtering rate was about 9 nm/min and sputtering time was set to be 100 seconds. Plasma enhanced chemical vapor deposition (PECVD) was used to grow Si3N4 protection layer at the top by using the oxford PECVD 80PLUS. The flow rates for the reactive gas of nitrogen, silane (5%) and ammonia, were set to be 600 sccm, 450 sccm and 20 sccm respectively with a chamber pressure of 650 mtorr. Besides, the sputtering temperature was 350° C and the sputtering rate was about 15 nm/min. The sputtering time was set to be 70 min, and the thickness of Si3N4 was about 1.0 μm.
The absorption layers were fabricated to stripe mesas of about 400 μm × 10 mm on MCNO film, with an interval of 0.6 mm by using wet etching techniques. Metallic electrodes of Ti/Au (30/150 nm) were deposited onto MCNO film by using UV lithography and double ion-beam sputtering. Then the MCNO devices were cracked into small pieces with the size of about 800 μm × 400 μm, where the sensitive area was about 400 μm × 400 μm. The detector A (without absorptive layer) was made of a bare MCNO film on sapphire, while the detector B was constructed with a sandwich structured absorption layer. The Current-Voltage (IV) characteristics of the detector B ranging 220-310 K were measured by using a Keithley 2400 Source meter at the DC mode. To conduct the noise spectra and frequency dependent response measurement, we used a low noise preamplifier ( × 360) together with a voltage amplifier ( × 5, SR 560, Stanford, USA), providing a bias voltage of 15 Volts with a total gain of 1800 times. In the noise spectra measurement, the voltage fluctuations of the detectors were measured by a SR770 (Stanford, USA) dynamic signal analyzer, where the detectors were blocked by a copper plate. The response waveforms to 1550 nm laser (MW-IR-1550, Changchun, China) of detectors A and B was conducted at a chopping rate of 5 Hz and 20 Hz, where no amplifier was needed because of the high power density of the laser. By using an OPHIR power meter, the incident power density on the detector for the 1550 nm laser source was measured and calculated by using the formula PT/ST, where ST is the area of the radiation beam, SD the active area, hence the incident power φs = AD∙PT/ST. The incident power density of 1550 nm in this study was estimated to be about 2.1 W/cm2. By using a 600 MHz oscilloscope (62Xi-A, Lecroy, USA), we measured the response signal waveforms of detectors at 5 Hz, which were used to determine the time constant of the detectors. To investigate the thermal sensing ability of the detectors, we conducted the blackbody measurement by exposing the detectors to a blackbody source (Electro Optical Industries, Inc., CS1050) at a temperature of 700 K. For the blackbody measurement, the incident power φs can be calculated by using the formula: where σ is the Stefan’s constant; Tbla, Tamb are the blackbody and ambient temperature; A0 and AD are the area of blackbody aperture (diameter 7 mm) and active area of the sensor element; L = 12 cm is the distance from the blackbody aperture to the detector. To conduct the relative response spectra measurement, we used a Bruker 80V Fourier transform spectrometer in the range of 400 to 7500 cm−1 by 32 scans with a scanning rate of 1.6 KHz, where a commercial DTGS pyroelectric detector served as a reference detector to measure the response spectra. The detailed experimental setup for response spectra measurement is shown in Fig. 1.
Fig. 1 The photo of a commercial DTGS detector (a) and the MCNO film detector (b), and the experimental setup for responsivity spectra measurement (c).
3. Results and discussions
The phase analysis about the structure of the MCNO sample was identified by the X-ray diffraction (XRD) result, as shown in Fig. 2(a). The sample exhibits multiple peaks, with cubic spinel phase with orientations mainly corresponding to (311) and (533) planes, which are indexed by making reference to NiMn2O4 with the JCPDS card No. 84-0542. The results of XRD indicates the polycrystalline nature of MCNO film. The cross-sectional image of MCNO film was examined by the scanning electron microscopy (SEM). As shown in the inset of Fig. 2(a), the MCNO film is densely formed without cracks or defects, with a thickness of about 10 μm.
Fig. 2 (a) XRD pattern of the MCNO film prepared by RF sputtering method on sapphire substrate. Inset: Cross-sectional picture of the MCNO films with thickness of about 10 μm. (b) Temperature dependent resistance and TCR coefficient of MCNO detector B. Inset: IV curves at the temperatures of 220 K, 240 K, 260 K, 280 K, 300 K and 310 K.
In this study, we measured the current-voltage characteristics of the sample B between 220 and 310 K. Temperature-dependent resistance (R-T) for the sample B was derived by fitting the experimental data by using a least square fitting method, and the temperature coefficient of resistance (TCR) curve can be derived from the R-T curves. Varied temperature resistance and TCR curves (R-T and TCR-T) of the sample B are shown in Fig. 2(b). As shown in Fig. 2(b), the resistance of the detector B is about 298 KΩ at 295 K (RT), and the TCR ranks about −4%K−1 to −6%K−1 in the range of 230 K to 310 K. The I-V characteristic curves of detector B are shown in the inset of Fig. 2(b), where the current is linearly dependent on voltage, indicating fine ohmic behavior. These results show that the MCNO film is suitable for developing thermal sensitive bolometers.
The absorption structure for the MCNO detector B is shown in Fig. 3(a). The absorption layer is coated on the thermal sensitive element and consists of a silicon nitride (Si3N4) dielectric layer, a nickel-chromium alloy (NiCr) absorption layer and an SiO2 insulating layer from top to bottom. In this study, a 100 nm silicon oxide (SiO2) film was coated onto MCNO film as an insulate layer between the thermal sensitive element and the metal absorptive layer. The NiCr film served as the IR absorption layer. Since the refraction index of NiCr layer is significantly higher than that of the vacuum, we used a dielectric layer of Si3N4 with a thickness of 1.0 μm as an anti-reflection (AR) layer and also a protective layer. The refractive index n_(Si3N4) is about 2.0 in the range of 3-14 μm. Therefore, the Si3N4 layer can also be regarded as a 1/4 waveplate AR film with a central wavelength of 8 μm, resulting in a reflection minus in the range of 5-10 μm. The schematic diagram of MCNO thin film detector B is shown in Fig. 3(b), where a detection unit and a compensator unit were adhered onto a copper heat sink by using the epoxy glue. The latter one is coated with silicone rubber to block the incident IR wave. The photos of the packaged MCNO detector and the optical images of MCNO sensitive element are shown in Fig. 3(c).
Fig. 3 (a) Sectional diagram of absorption structure for the MCNO detector. From top to bottom: the protective layer (Si3N4), the absorption layer (NiCr), the insulate layer (SiO2), the thermal sensitive material (MCNO), and the substrate (sapphire). (b) The schematic diagram of MCNO detector with the absorption layer. (c) A photo of the packaged MCNO detector and optical images of MCNO sensitive element.
The IR absorption properties of the MCNO film are critical parameters in designing MCNO IR thermal detectors. According to the formula:
we can calculate the total absorption of the two samples, as shown in Fig. 4(c), where R is the reflectivity, T is the transmittance, and Abs is the absorption. As shown in Fig. 4(a), the reflectance and transmittance of samples of sapphire-MCNO films with or without the sandwich absorption structures are measured in the range of 400-7500 cm−1. For comparison, the transmission (T%) spectra of sapphire substrate and T% spectra of sample A and B are shown in Fig. 4(b), where T% spectra of sample B significantly decreases in the range of 1.33 to 8 μm in comparison to that of sample A. In the range above 10 μm, the sapphire substrate is opaque and contributes to part of absorption. As reported in former works, the RF sputtered polycrystalline MCNO thin films is nearly transparent in the range of 3-14 μm (k~0) with a refractive index n of about 2.2-2.4, which is convinced by the multiple fringes of the reflecting spectra in Fig. 4(a) [17]. The IR absorption mechanism of the detector B is complex, which can be explained as follows: Firstly, in the range of 1.33-2.5 μm (4000-7500 cm−1), the incident wave is majorly absorbed by the MCNO film, where band to band transition exists above 0.5 eV (<2.5 μm) [14,21]. Secondly, in the range of 2.5-7.5 μm (1300-4000 cm−1), both of the MCNO film and sapphire substrate are transparent, thus, the absorption reaches the minimum, while the 50% absorption of detector B should be attributed to the free carrier absorption in the NiCr metal film. Thirdly, in the range of 7.5-10 μm (1000-1300 cm−1), the IR vibration of Al-O bond in sapphire substrate contributes to part of the absorption [22]. Fourthly, in the range of 10-14 μm (700-1000 cm−1), the MCNO film is almost a transparent spacer, while the sapphire substrate acts as a highly reflective mirror. Therefore, the transmitted IR wave will be reflected at the MCNO-sapphire interface and absorbed by the NiCr layer twice. Lastly, the IR vibration mode in MCNO film contributes to phonon absorption in the range of 14-25 μm (400-700 cm−1) [17], where the extinction coefficient k ranks above 1, leading to 50-60% reflection in sample A. In comparison the AR film Si3N4 works in this wavelength range and reduces the reflection to be less than 20%. To summarize, benefited from the sandwich structured absorptive layer, the MCNO detector B achieved an average absorption of about 50% to 90% in the wavelength range of 1.33-25 μm.
Fig. 4 (a) Comparison of T% and R% of samples A and B. (b) Comparison of T% of samples A and B and the sapphire substrate. (c) Absorption spectra of sample A and B.
To evaluate the performance of the detectors, we measured and compared the waveforms, frequency-dependent response curves, as well as the response spectra of A and B. As shown in Fig. 5(a), the measured time constants of the 0.1 Vpp to 0.9 Vpp for detectors A and B to a 1550 nm laser were approximately 13.0 ms and 14.5 ms. As a type of thermal detection, the response time (τ), responsivity (RV) and detectivity (D*) of the MCNO detector can be estimated as:
Fig. 5 (a) Response waveforms of MCNO thin film detectors to the 1550 nm laser (@5 Hz, 20 Hz). (b) Frequency dependent noise of MCNO thin film detector A and the responsivity values of A and B to a 700 K blackbody.
where VS is extracted from the input value of lock-in amplifier, φs is the incident power received by the detector, AD is the active area of the detector, and NV the voltage fluctuation spectral density. Here, for the 1550 nm laser measurement, the incident power φs is calculated to be about 3.36 mW, and the RV of detectors A and B to a 1550 nm is calculated to be 70.4 V/W@20 Hz and 98.6 V/W@20 Hz, respectively.
For the blackbody measurement, the incident power φs can be calculated according to the Stefan’s law. Figure 5(b) shows the frequency dependent response of MCNO thin film detectors to a 700 K blackbody, where the time constant can be calculated according to the formula [8]:
where the factor η represents the absorption of the detector, Vb the bias voltage of the detector, ω the circular frequency, τ the time constant, G the thermal conductance. According to Formula (4), the time constants are fitted to be 22.4±2.0 ms (A) and 26.7±1.6 ms (B), respectively, which are about the same order with the values derived from the waveforms in Fig. 5(a). As shown in Fig. 5(b), the detector B showed a responsivity of 38.0 V/W@20 Hz and D* of 0.81 × 107 cmHz0.5/W@20 Hz to a 700 K blackbody, which increased by about 70% in comparison to that of the detector A (22.2 V/W@20 Hz).
The response spectra of the detectors A and B to a Globar IR source were shown in Fig. 6(a), and the relative response can be determined by using the response spectra of a commercial DTGS as the reference detector. Here, the spectral response of the commercial DTGS is considered to be flat since its detection mechanism is non frequency selective, and its response to the Globar source is also shown in Fig. 6(a). According to the RV values measured at 1550 nm, the responsivity spectra of the detectors A and B in the whole range can also be figured out. As shown in Fig. 6(b), the response spectra of the detector B cover a broad range of 1.33 to 25 μm, with a small increase of about 10% to 20% in time constants. The responsivity of detector B to a 700 K blackbody reaches about 80 V/W@ 20 Hz, 5 μm, which is enhanced by 400% in comparison to that of the detector A (17 V/W@20 Hz, 5 μm), and the D* value of the detector B reaches 1.7 × 107 cm∙Hz0.5/W@20 Hz, 5 μm. The fluctuations of the responsivity at different wavelength range may originate from the different photo-thermal process and thermal conducting process, as analyzed in former paragraphs. The experimental results are summarized below in Table 1. Since no thermal isolation structure was applied in this work, the thermal conductance of the detector is about 2-3 orders larger than that of the thermal isolated microbolometers. The performance of a micro bridge structured MCNO bolometer was evaluated according to the formula (4) in former works, which can reach as high as 0.6 × 109 cm∙Hz0.5/W@30 Hz [20].
Fig. 6 (a) Response spectra of the detectors to a Globar IR source. Inset: interference signal of the detector B for the IR source. (b) Comparison of the relative response spectra of A and B in the range of 1.33 to 25 μm.
Table 1. Summary of Results for Two MCNO Thin Film Detectors (Blk: Blackbody)
4. Conclusion
We have designed and fabricated a dielectric-metal-dielectric structured absorption layer for MCNO film detectors for the first time, which has improved the absorption in the 3-14 μm IR transparent range of MCNO film. Among them, the absorption of 3-7.5 μm wave should be attributed to the NiCr layer, and the 7.5-10 μm wave is absorbed by both the sapphire substrate and the NiCr layer, then the 10-14 μm absorption is majorly absorbed by the sapphire-MCNO-NiCr cavity. The responsivity of the detector B has been greatly improved by 70% to a 700 K blackbody in comparison to that of the detector A, and the D* of detector B reached 1.7 × 107 cm∙Hz0.5/W@20 Hz, 5 μm. The response spectra of B covers a broad range of 1.33 to 25 μm, with a small increase in time constant of merely 10% to 20%. Our experimental results are of great significance for the future development of broad band uncooled infrared array MCNO detectors.
Funding
China National Funds for Distinguished Young Scientists (61625505); National Natural Science Foundation of China (61604160); Natural Science Foundation of Shanghai (17ZR1444100); Innovation Shanghai Institute of Technical Physics (CX-197).
References
1. N. Pahlevan, J. C. Roger, and Z. Ahmad, “Revisiting short-wave-infrared (SWIR) bands for atmospheric correction in coastal waters,” Opt. Express 25(6), 6015–6035 (2017). [CrossRef] [PubMed]
2. Y. Wang, M. A. Ali, E. K. C. Chow, L. Dong, and M. Lu, “An optofluidic metasurface for lateral flow-through detection of breast cancer biomarker,” Biosens. Bioelectron. 107(1), 224–229 (2018). [CrossRef] [PubMed]
3. X. Chong, E. Li, K. Squire, and A. X. Wang, “On-chip near-infrared spectroscopy of CO2 using high resolution plasmonic filter array,” Appl. Phys. Lett. 108(22), 221106 (2016). [CrossRef]
4. C. H. Chen, X. J. Yi, X. R. Zhao, and B. F. Xiong, “Characterizations of VO2-based uncooled microbolometer linear array,” Sens. Actuators A Phys. 90(3), 212–214 (2001). [CrossRef]
5. Li, C., G. D. Skidmore, and C. J. Han,”Uncooled vox infrared sensor development and application,” Infrared Technology and Applications Xxxvii 8012 (2011).
6. A. Lisauskas, S. I. Khartsev, and A. Grishin, “Tailoring the colossal magnetoresistivity: La0.7(Pb0.63Sr0.37)0.3MnO3 thin-film uncooled bolometer,” Appl. Phys. Lett. 77(5), 756–758 (2000). [CrossRef]
7. J. H. Kim, S. I. Khartsev, and A. M. Grishin, “Epitaxial colossal magnetoresistive La-0.67(Sr,Ca)(0.33)MnO3 films on Si,” Appl. Phys. Lett. 82(24), 4295–4297 (2003). [CrossRef]
8. C. OuYang, W. Zhou, J. Wu, Y. Hou, Y. Q. Gao, and Z. M. Huang, “Uncooled bolometer based on Mn1.56Co0.96Ni0.48O4 thin films for infrared detection and thermal imaging,” Appl. Phys. Lett. 105(2), 22105 (2014). [CrossRef]
9. Y. Hou, Z. M. Huang, Y. Q. Gao, Y. J. Ge, J. Wu, and J. H. Chu, “Characterization of Mn(1.56)Co(0.96)Ni(0.48)O(4) films for infrared detection,” Appl. Phys. Lett. 92(20), 202115 (2008). [CrossRef]
10. R. Schmidt and A. W. Brinkman, “Studies of the temperature and frequency dependent impedance of an electroceramic functional oxide NTC thermistor,” Adv. Funct. Mater. 17(16), 3170–3174 (2007). [CrossRef]
11. C. H. Chen, X. J. Yi, and B. F. Xiong, “Infrared responsivity of uncooled VO2-based thin films bolometer,” Wuli Xuebao 50(3), 450–452 (2001).
12. S. Baliga, A. L. Jain, and W. Zachofsky, “Sputter Deposition and Characterization of Ni-Mn-O and Ni-Co-Mn-O Spinels on Polymide and Glass Substrates,” Appl. Phys., A Mater. Sci. Process. 50(5), 473–477 (1990). [CrossRef]
13. G. Ji, A. M. Chang, J. B. Xu, H. M. Zhang, J. Hou, B. Zhang, and P. J. Zhao, “is Low-temperature (<300°C) growth and characterization of single-[100]-oriented Mn–Co–Ni–O thin films,” Mater. Lett. 107, 103–106 (2013). [CrossRef]
14. W. Zhou, L. B. Zhang, C. Ouyang, J. Wu, Z. M. Huang, and X. F. Xu, “Micro structural, electrical and optical properties of highly (220) oriented spinel Mn-Co-Ni-O film grown by radio frequency magnetron sputtering,” Appl. Surf. Sci. 311, 443–447 (2014). [CrossRef]
15. Q. Shi, W. Ren, X. B. Zhang, W. W. Kong, B. Gao, L. Wang, J. B. Xu, A. M. Chang, and L. Bian, “Effect of Ar/O2 ratio on structure and cationic distribution of Mn1.56Co0.96Ni0.48O4+/−delta spinel films,” Appl. Surf. Sci. 405, 47–51 (2017). [CrossRef]
16. W. Zhou, J. Wu, C. Ouyang, Y. Q. Gao, X. F. Xu, and Z. M. Huang, “Optical properties of Mn-Co-Ni-O thin films prepared by radio frequency sputtering deposition,” J. Appl. Phys. 115(9), 93512 (2014). [CrossRef]
17. R. Dannenberg, S. Baliga, R. J. Gambino, A. H. King, and A. P. Doctor, “Infrared optical properties of Mn1.56Co0.96Ni0.48O4 spinel films sputter deposited in an oxygen partial pressure series,” J. Appl. Phys. 86(5), 2590–2601 (1999). [CrossRef]
18. E. M. Smith, D. Panjwani, J. Ginn, A. P. Warren, C. Long, P. Figuieredo, C. Smith, J. Nath, J. Perlstein, N. Walter, C. Hirschmug, R. E. Peale, and D. Shelton, “Dual band sensitivity enhancements of a VOX microbolometer array using a patterned gold black absorber,” Appl. Opt. 55(8), 1 (2016).
19. S. Karanth, M. A. Sumesh, V. Shobha, H. Ganesh Shanbhogue, C. L. Nagendra, “Infrared detectors based on thin film thermistor of ternary Mn-Ni-Co-O on micro-machined thermal isolation structure”, Sensors and Actuators a-Physical 153(1), 69 (2018).
20. W. Zhou, Y. M. Yin, J. Wu, Y. Q. Gao, and Z. M. Huang, “Improvements in electrical properties, low frequency noise and detection performance of a Mn-based bilayer thin film infrared detector,” Sens. Actuators A Phys. 283(1), 196–203 (2018). [CrossRef]
21. J. Wu, C. Ouyang, W. Zhou, Y. Q. Gao, Y. Hou, Z. M. Huang, and J. H. Chu, “Optical properties of small polaron in Mn1.56Co0.96Ni0.48O4 films investigated by temperature-dependent transmission spectra,” Mater. Res. Express 2(2), 26301 (2015). [CrossRef]
22. E. D. Palik, Handbook of Optical Constants of Solids, 2nd edition (1998).
