An emissivity measurement apparatus for near infrared spectrum

An emissivity measurement apparatus for near infrared spectrum

Infrared Physics & Technology 73 (2015) 275–280 Contents lists available at ScienceDirect Infrared Physics & Technology journal homepage: www.elsevi...

721KB Sizes 6 Downloads 152 Views

Infrared Physics & Technology 73 (2015) 275–280

Contents lists available at ScienceDirect

Infrared Physics & Technology journal homepage: www.elsevier.com/locate/infrared

An emissivity measurement apparatus for near infrared spectrum Feng Zhang a, Kun Yu b,⇑, Kaihua Zhang b, Yanlei Liu a, Kaipin Xu b, Yufang Liu a,b,⇑ a b

School of Optoelectronics, Beijing Institute of Technology, Beijing 100081, China College of Physics and Electronic Engineering, Henan Normal University, Xinxiang 453007, China

h i g h l i g h t s  A new emissivity measurement apparatus for near infrared spectrum is reported.  We present an improved method to minimize the measurement error.  The spectral emissivities of TA1, oxidized nickel and 304 austenitic stainless steel are measured and discussed.  The uncertainty of the apparatus is discussed and calculated in detailed.

a r t i c l e

i n f o

Article history: Received 26 February 2015 Available online 20 October 2015 Keywords: Spectral emissivity Experimental apparatus Near-infrared spectrum Uncertainty

a b s t r a c t This study develops a new experimental apparatus for infrared spectral emissivity measurements which consists mainly of the following four parts: sample heating system, blackbody furnace, optical system, and data acquisition system. This apparatus focuses on the near-infrared spectral emissivity measurement covering the temperature range from 473 K to 1273 K and the wavelengths between 0.8 lm and 2.2 lm. The apparatus and the measurement method are described in detail, and an improved method is presented to minimize measurement error. The spectral emissivity of pure titanium TA1, oxidized nickel and 304 austenitic stainless steel are measured to validate the reliability and reproducibility of experimental apparatus. The experimental results in this study are in good agreement with those of other literatures. Various uncertainty sources in emissivity measurement are analyzed, and the combined standard uncertainty of this system is less than 3.9%. Ó 2015 Elsevier B.V. All rights reserved.

1. Introduction The infrared emissivity is a key material parameter of scientific implication and application in such areas as industrial production, aerospace industry, scientific research, radiation thermometry, telemetry and infrared guiding technique [1,2]. For example, in radiation thermometry, a 0.1 uncertainty in emissivity of a sample at 1273 K may correspond to a temperature error of about 35 K if the emissivity value is about 0.9, and the temperature error will reach 135 K if the emissivity value is about 0.2 [3]. The emissivity of materials is influenced by numerous factors including temperature, measuring wavelength, chemical composition, surface roughness and surface oxidation [4]. Even for the same material, the values of emissivity described in different literatures are sometimes different from each other. Therefore, it is necessary to

⇑ Corresponding author at: School of Optoelectronics, Beijing Institute of Technology, Beijing 100081, China. Tel.: +86 373 3326187 (Y. Liu). E-mail addresses: [email protected] (K. Yu), [email protected] (Y. Liu). http://dx.doi.org/10.1016/j.infrared.2015.10.001 1350-4495/Ó 2015 Elsevier B.V. All rights reserved.

investigate further the emissivity of materials and evolve with the improvement of measuring apparatus. Many facilities about emissivity measurement have been described in the literature over the past decades [5–9]. The methods of these facilities can be broadly classified into two categories: indirect and direct emissivity measuring methods. Indirect measuring refers mainly to the reflection method [9] in which the emissivity can be calculated based on the Kirchhoff’s law by measuring the reflectivity and the transmissivity of the sample [10]. Direct emissivity measuring method is based on the definition of emissivity by measuring the radiance of the sample and blackbody at the same conditions (e.g. temperature, wavelength and angle), in which the emissivity is computed as the ratio of the two measuring values. At present, direct emissivity measuring method often employs Fourier transform infrared spectrometer (FTIR) and monochromator. The former has such advantages as fast measurement and wide wavelength range [12], but the expensive cost makes it difficult to be introduced into general use. The latter is now widely employed to establish high-precision emissivity measurement systems owing to its advantages such as simple

276

F. Zhang et al. / Infrared Physics & Technology 73 (2015) 275–280

and accurate principle, low cost, narrow width and high accuracy wavelength positioning [11,13–15]. The emissivity in near-infrared spectral range is commonly used in radiation pyrometers and fiber-optic colorimeters, and the reported data about this wavelength band is in shortage. Hence, it is significant to investigate the emissivity behavior of materials to enrich the emissivity data in this wavelength band. A new experimental apparatus is developed to provide accurate values of emissivity in near-infrared spectral ranges from 0.8 lm and 2.2 lm at the temperature between 473 K and 1273 K. A high-accuracy grating monochromator, a professional detector for near-infrared spectral range and a lock-in amplifier are employed to measure the normal spectral emissivity for solid materials accurately. Furthermore, in the past decades, most of the emissivity measurements were performed in the vacuum condition. But it is impossible to perform the radiation thermometry or other operations in such an ideal condition in practical use. For this reason, the effect of surface oxidation on the spectral emissivity needs to be taken into account, and the measurements are performed in atmospheric conditions to meet the operational conditions of interest in this research. 2. Experimental setup The experimental apparatus as shown in Fig. 1 consists of the following parts: sample heating and temperature control system, blackbody furnace (ISOTECH R970), grating monochromator (omni-k5007), lock-in amplifier (SR830) and optical chopper (SR540), two off-axis parabolic gold-coated mirrors, and motorized rotation stage. Radiation from the sample or blackbody is achromatically reflected from the off-axis parabolic gold-coated mirror 3 into a collimated beam. The collimated beam is then focused by another off-axis parabolic protected gold-coated mirror 4 and directed into the entrance slit of grating monochromator. Before reaching the entrance, the beam is modulated to an appropriate frequency by the optical chopper system to distinguish the intensity signal from background noise. In addition, a SD six-position optical filter wheel is installed in front of the entrance slit of grating monochromator to eliminate the impacts of secondary spectrum and stray radiation. Then the output signal is detected by a high performance InGaAs infrared detector at the spectral range between 0.8 lm and 2.2 lm, and the signal of the InGaAs detector is amplified by a lock-in amplifier and fed to data capture system. This design can effectively eliminate the noise signal and improve the accuracy of measurement.

2.1. Blackbody An ideal blackbody is defined as a source with the emissivity of 1 at all wavelengths and temperatures. However, the emissivity of actual blackbody is always less than 1. In order to meet the requirements of this system, the blackbody with high and stable emissivity should be chosen. The blackbody furnace ISOTECH R970 is employed, whose effective emissivity is higher than 0.995 with the reference temperature ranging from 423 K to 1473 K. As shown in Fig. 2, the cavity of the furnace with the diameter being 20 mm is removable, and a fixed point cell can be put in its place to calibrate the temperature measuring equipment. The cavity inside the fixed point cell is 10 mm in diameter by 65 mm deep to the tip of a 120° cone. The temperature of the furnace is set on a controller, while the independent indicator indicates the actual radiance temperature using a high performance type R thermometer. 2.2. Sample heating system The self-designed sample heating system is mainly composed of a cast iron plate heater and a temperature control system. As shown in Fig. 3 the electric heater of resistance wire is spirally inserted in the cast iron plate. The specimen inside the chamber is heated by the iron plate heater. To minimize the heat loss and shield the noise from the surrounding environment, the whole heater is highly insulated using aluminum silicate wool and enclosed in a stainless steel shell. The temperature control system is consisted with two main modules: a high resolution and precision digital PID (Shimaden SR23) controller and a highperformance power regulator. The temperatures of the samples are monitored by two high accuracy type K thermometers which are symmetrically-welded near the measuring point of the sample surface by using thermocouple welder. The output of one thermocouple as input signal is fed back to the digital PID controller and then calculated using intelligence PID control algorithm based on the expert system, and the output signal of another thermocouple is fed back to a temperature acquisition unit to monitor the sample temperature. The PID controller outputs a signal to control the power regulator (SHIMADEN PAC01A), letting the iron plate heater cast a suitable heating power. This design can greatly improve the temperature control precision and the heating efficiency. The temperature range is from 473 K up to 1273 K and the control precision is 0.1 K. Samples were processed into disc with the diameter being about 50 mm and the thickness being 2 mm. A disc aluminum silicate insulation board is mounted in front of the measuring sample to reduce the Size of Source Effect (SSE), which represents the relative contribution to the measure signal due to radiation outside the nominal measured area of the sample. 2.3. The optical system

Fig. 1. Schematic diagram of the spectral emissivity measurement apparatus. 1 – Blackbody, 2 – sample furnace, 3 and 4 – parabolic gold-coated mirrors, 5 – motorized rotation stage, 6 – filter wheel, 7 – chopper, 8 – grating monochromator, 9 – detector, 10 – lock-in amplifier, and 11 – data capture system.

The optical system of the apparatus consists of two parabolic mirrors, a motorized rotation stage, a grating monochromator and an optical table. To minimize the aberrations and reduce the loss of the radiation, two off-axis parabolic gold-coated mirrors are employed which can collimate a point light source or focus into a collimated beam. The effective focal length is 203.2 mm and the average reflectance is over 96% at a broadband range from 0.8 lm to 20 lm. The all-reflective design eliminates phase delays and absorption loss introduced by transmissive optics. The off-axis parabolic gold-coated mirror 3 is mounted on the motorized rotation stage, and the other off-axis parabolic mirror 4 is installed on optical table by an optical frame, the height of which could be adjusted precisely. The cast iron plate heater and the blackbody furnace are symmetrically-arranged on the two sides of off-axis

F. Zhang et al. / Infrared Physics & Technology 73 (2015) 275–280

277

Fig. 2. Diagram of the black body furnace.

Fig. 3. Schematic of the cast iron plate heater.

parabolic gold-coated mirror 3, and their measurement points are both regulated to the focus of mirror 3. The motorized rotation stage with high reliable accuracy of rotation for its worm and worm wheel has adopted durable friction material, which rotates and drives mirror 3 to rotate to the specified angles, and finishes angle adjustment automatically. The maximum speed of motorized rotation stage is 25°/s with its repeatability being 0.005°. As the core component of this apparatus, the grating monochromator (Omni-k5007) is equipped with three gratings and two exit slits to enable simultaneous mountings of different detectors. In this arrangement, the monochromator can continuously spite light for the wavelengths range from 200 nm to 2500 nm. The incident slit is equipped with a special set of order-sorting filter wheels to block higher orders of the diffracted light. When gratings are revolving anti-clockwise, the lights from the exit slit are listed in the order of wavelength. Furthermore, all mirrors and grids used in the monochromator are gold coated to obtain a high throughput. The alignment of the optical system has a vital influence on the measurement results. The radiation emitted from the radiation source (sample or blackbody) should be effectively directed and focused into the monochromator to ensure the precise measurement because of the weak radiation intensity and high noise

strength in the near infrared wavelength range. If the radiation emitted by the sample is not focused into the monochromator perfectly, the effective radiation tested by the detector will be smaller than actual value, thus the measured emissivity value will be smaller than the real value. Rather, the measured emissivity value will be bigger than the real one when the radiation of the blackbody is not directed into the monochromator effectively. And the above two cases will lead to effective signal reduction and strong noise enhancement. As we know, the emissivity value varies with the change of measurement angles, so the optic axis of mirror must be perpendicular to the sample strictly to accurately measure the normal emissivity. The optical path from both sample and blackbody to the detector must be completely equal to each other to reduce the error caused by the radiation loss in the air. A visible point light source is employed for the optical system alignment, which is fitted on the entrance of the monochromator. Mirror 4 is put against the entrance at a distance of focal length. Then a CCD matrix camera is put between mirrors 3 and 4 to adjust the position of mirror 4 making sure the emergent light is parallel. Then the CCD camera is put in front of the sample chamber or blackbody to adjust the position of mirror 3, so that the emergent light of mirror 3 be focused on the sample or blackbody. 2.4. Data capture system The TE InGaAs detector, which offers high sensitivity in near infrared spectral range, is employed in data capture system to improve the accuracy of the measurement. The spectral range of the TE cooled InGaAs detector is from 0.8 lm to 2.2 lm, and the detector need work with temperature controller (ZTC) for cooling control at 40 °C by using thermoelectric refrigeration. Furthermore, the temperature controller needs to be stabilized for several minutes before each test in order to avoid errors caused by temperature variations. The signals of the InGaAs detector are amplified and filtered by lock-in amplifier (SR830). The lock-in amplifier uses a technique of phase-sensitive detection to single out the component of the signal at a specific reference frequency and phase. Noise signals at nonreference frequencies are rejected without affecting the measurement results. 3. Experimental principle The direct measurement method is based on the emissivity definition, which is the ratio of the radiance emitted by sample

278

F. Zhang et al. / Infrared Physics & Technology 73 (2015) 275–280

surface against that emitted by blackbody at the same wavelength and temperature. The spectral radiance from the sample surface can be written as follows:

Leff ðk; T s Þ ¼ es ðk; T s ÞLb ðk; T s Þ þ ½1  es ðk; T s ÞLb ðk; T am Þ

ð1Þ

where eðk; T s Þ is the emissivity of the sample at temperature Ts, Lb ðk; T s Þ is the spectral radiance of the blackbody at temperature Ts, and Lb ðk; T am Þ is the spectral radiation from the surrounding at ambient temperature Tam. As we know, the ambient environment can be regarded as a blackbody [1], and the chopper is black painted with high emissivity coating with its emissivity higher than 0.95. So, the effective spectral radiance from the optical chopper surface is very close to that from the blackbody at the same ambient temperature. The electrical signals from the InGaAs detector reflect the difference between the effective spectral radiance from the sample and the optical chopper surface, which can be written as:

Is ¼ K es ðk; T S Þ½Lb ðk; T s Þ  Lb ðk; T am ÞgðkÞRðkÞDk

ð2Þ

where K is the coefficient of the apparatus related to the collection parameter of the measurement system, RðkÞ is the spectral responsibility of the InGaAs, gðkÞ is the spectral efficiency of the grating monochromator, and Dk is the wavelength bandwidth. Analogously, the electrical signals from the blackbody can be written as:

Ib ¼ K eb ðk; T b Þ½Lb ðk; T b Þ  Lb ðk; T am ÞgðkÞRðkÞDk

ð3Þ

Then the emissivity of the sample is given by

es ðk; T s Þ ¼ eb ðk; T b Þ

IS Lb ðk; T b Þ  Lb ðk; T am Þ Ib Lb ðk; T s Þ  Lb ðk; T am Þ

ð4Þ

Eq. (4) is the basic calculation equation of the emissivity measuring system. If the temperatures of the sample and the blackbody are assumed to be identical, Eq. (2) can then be simplified as follows:

es ðk; T s Þ ¼ eb ðk; T b Þ

IS Ib

ð5Þ

Independent of the surrounding temperature, the emissivity of the sample at this point is determined only by two elements: the effective emissivity of the blackbody and the spectral corresponding current radio of sample to the blackbody. In the application, it is very difficult to measure the true temperature of the sample surface and thus very difficult to ensure a consistent temperature with the blackbody. Before the establishment of the measurement apparatus, the thermocouples of the sample and blackbody have been calibrated by fixed-point blackbodies, and the errors of these thermocouples in different temperature ranges have been determined. The temperature range of this experiment is from 473 K to 1273 K. The maximum temperature drift of the K-type thermocouples, which are welded on the sample surface, is 1.1 K at 473 K and 1.5 K at 1273 K. The maximum temperature drift of the R-type thermocouple, which is mounted in the blackbody, is 0.2 K at 473 K and 0.4 K at 1273 K. The different error compensations have been added to the computation to minimize the temperature difference between the sample and blackbody. But the small temperature difference between the sample surface and blackbody still occurs which would cause a measurement error. In order to improve the measuring precision, an improved method is presented in this study by introducing a corrected item nðkÞ. By definition, the spectral emissivity of the sample at temperature Ts can be written as follow:

eðk; T S Þ ¼

LS ðk; T S Þ Lb ðk; T S Þ

The ratio of practical measured signals is given by

nðkÞ ¼

LS ðk; T S Þ Lb ðk; T b Þ

ð7Þ

Solving Eq. (7) for LS ðk; T S Þ, the radiance emitted by the sample can be described as a function of the signals ratio nðkÞ and the signal of blackbody radiance Lb ðk; T b Þ.

LS ðk; T S Þ ¼ nðkÞLb ðk; T b Þ

ð8Þ

Then, the emissivity can be accurately calculated by combining Eqs. (6) and (8) as follows:

eðk; T S Þ ¼ nðkÞ

Lb ðk; T b Þ Lb ðk; T S Þ

ð9Þ

4. Measurements and results 4.1. Emissivity of pure titanium In this research, the titanium (TA1) is chosen to validate spectral emissivity measurement system because of its advantages such as good thermo stability, anticorrosion and compressive strength, and the measurements are carried out at temperatures of 473 K, 573 K, 673 K and 773 K with spectral range from 0.8 lm to 2.2 lm. The titanium samples are processed into the required size and polished with SiC abrasive paper (1000 grits) to ensure a specified roughness. The polished samples are cleaned in succession with ethanol and acetone to remove the oil, grease or dirt. Then, the surface roughness of samples was measured using a roughness tester (Time TR220) and the final roughness average Ra is 0.078 ± 0.003 lm. The roughness parameters of the sample at the four temperatures are very close to each other, so the influence of the roughness can be ignored [17,18]. For each sample, the same measurement is performed for five times at each temperature, and an excellent reproducibility is observed in the measurement procedure. Emissivity of the samples at 473 K, 573 K, 673 K, 773 K in the wavelengths ranging from 0.8 lm and 2.2 lm is calculated and shown in Fig. 4. The experimental results show that the emissivity of pure titanium TA1 increases with the rise of the temperature. In the wavelength range between 0.8 lm and 2.2 lm, the emissivity decreases linearly and slowly with the increase of the wavelength at given temperature according to the electromagnetic theory for metallic materials [19]. The measurement results are consistent with the general rule of metal emissivity indicating that the TA1 samples do not have the obvious oxidizing reaction in the measuring process. These results accord well with those reported by Teodorescu G in literature [16] at the wavelength range of 0.8–2.2 lm. For instance, the emissivity

ð6Þ Fig. 4. Spectral emissivity of TA1 at 473 K, 573 K, 673 K and 773 K.

279

F. Zhang et al. / Infrared Physics & Technology 73 (2015) 275–280

in air environment to reach a similar surface oxidation degree to the literature [20]. The spectral emissivity measurements of the two oxidized Ni samples are carried out at 640 K and 758 K in the wavelength range from 0.8 lm to 2.2 lm, the results of which are shown in Fig. 5. The spectral emissivity values decrease with the increase of wavelength at a given temperature, and the emissivity values at 640 K are bigger than those at 758 K. The results accord well with those reported by Hanssen [20]. For example, the emissivity value at 758 with wavelength 2.0 lm is 0.81 in this study, and in the literature of Hanssen, the emissivity value is 0.82 under this condition [20]. 4.3. Emissivity of stainless steel

Fig. 5. Spectral emissivity of oxidized nickel.

Two 304 austenitic stainless steel samples (Nos. 1 and 2) are prepared. The samples are polished with 1200 grit SiC papers and then polished with 0.5 lm silica colloidal solution. The polished samples 1 and 2 are exposed at 773 K and 973 K in air for 1 h respectively to get a same surface condition with the literature [21], then the emissivity of the samples are measured. As shown in Fig. 6, a peak is observed on the emissivity curve in wavelength 0.8–1.0 lm. This phenomenon can be explained based on interference principle [22]. The light path of the light reflected at the interface between atmosphere and oxide film is relatively different to that of the light reflected at the interface between oxide film and metal substrate, thus the constructive interferences and the destructive interferences could occur based the interference conditions. The emissivity values of sample 1 are in the range of 0.42–0.45 for wavelengths from 0.8 to 2.2 lm, and those of sample 2 are in the range of 0.58–0.68 which accord well with the results reported by Cao [21]. 5. Measurement uncertainty

Fig. 6. Spectral emissivity of 304 austenitic stainless steel.

value of 0.445 is measured at 773 K and 1.5 lm in this study, and in the article of Teodorescu G, the emissivity value under this condition is 0.449 [16]. When compared with the measuring results reported by Teodorescu G, some small differences are observed, which could result from the surface roughness of samples as it has remarkable effect on the emissivity in the near-infrared waveband. 4.2. Emissivity of oxidized nickel Two pure nickel samples are prepared identical to those of TA1 samples. Then the Ni samples are oxidized at 773 K for two hours

Measuring the emissivity requires the measurement of many parameters and the analysis of all the uncertainty sources [23,24]. The uncertainty of the emissivity measurement apparatus is caused by such components as determining wavelength, temperatures of the sample and blackbody, ambient temperature, emissivity of the blackbody, wavelength precision of the monochromator, noise on the measure spectra, etc. Furthermore, the combined standard uncertainty varies with the change of wavelength and temperature. The uncertainty increases considerably for low emissivity values at low temperatures and short wavelengths [25]. The uncertainties of the assembly unit used in this system have been evaluated before the setting up of the apparatus, and the key components of the measurement apparatus have been optimized. The uncertainty parameters of this apparatus are compared with those reported by other literatures [15,26] in Table 1. Uncertainties of the blackbody emissivity mainly originate from the difference between the effective emissivity of the cavity and

Table 1 Uncertainty estimates of the emissivity measurement. Parameter

473 K 0.8 lm e = 0.432

773 K 2.2 lm e = 0.427

Uncertainty of blackbody emissivity Blackbody temperature Sample temperature Stability of the BB temperature Stability of the sample temperature Nonlinearity of monochromator Noise on the measure spectra Wavelength precision of the monochromator(nm)

0.0025 0.2 K 1.1 K 0.1 K 0.1 K 0.1 0.43 0.024

0.0025 0.4 K 1.5 K 0.1 K 0.1 K 0.1 0.17 0.011

Total uncertainty in emissivity

3.90%

1.50%

Teodorescu et al. [16,26] e = 0.531, T = 1088 °C 0.30% 0.50% 0.04%

4%

Hatzl et al. [15] 0.0041 3.4 K 6K

5.10%

280

F. Zhang et al. / Infrared Physics & Technology 73 (2015) 275–280

the ideal value. Calculating the effective emissivity of the blackbody used in this system based on the Monte Carlo method [27], the uncertainty of blackbody emissivity is 0.0025 in the temperature range from 473 K to 973 K. The precision of the sample and blackbody temperature determination is better than that reported in previous literatures because of the calibration of thermocouples, error compensation and the improvement temperature calculation method [15,26]. Furthermore, the self-design temperature control system leads the temperature stability of the sample and blackbody to a higher level. According to the verification certificate of the monochromator (Omni-k5007), the nonlinearity of monochromator is 0.10, and the wavelength precision at 0.8 lm and 2.2 lm is 0.024, 0.011 respectively. For the every sample measurement, we averaged the results of five individual series of measurements and the monochromater repeated 20 scans in each measurement, then we calculate the noise on the emissivity spectra which influence the emissivity values and wavelength strongly. Through analyzing and calculating all the uncertainty sources shown in Table 1, the combined uncertainty in emissivity measurement of TA1 samples at T = 473 K, k = 800 nm is 3.9%, and that at T = 773 K, k = 2200 nm is 1.5%. According to the principle of uncertainty [25], the combined standard uncertainty of the experimental apparatus is less than 3.9%. 6. Summary In this paper, an emissivity measurement apparatus of opaque solid materials is developed at temperatures from 473 K to 1273 K and wavelengths ranging from 0.8 lm and 2.2 lm by using the grating monochromator, and an improved method is presented to minimize measurement error. The spectral emissivity measurements of pure titanium TA1, oxidized nickel and 304 austenitic stainless steel are performed at different temperatures and different surface conditions, and the data trend is well consistent with the general rule of metal emissivity. The measurement results are in good agreement with the data reported by other literatures at the wavelength range of 0.8–2.2 lm, which shows that the spectral emissivity measurement system is a reliable apparatus in measuring the emissivity of materials in near-infrared spectral range. Various uncertainty sources in emissivity measurement are considered, and the combined standard uncertainty of this system is less than 3.9% for the wavelength range 0.8–2.2 lm. This experimental apparatus enables us to investigate the emissivity behavior of materials in atmospheric condition and significantly improves the accuracy of emissivity measurement in near-infrared spectral range. Conflict of interest There is no conflict of interest statement. Acknowledgements This research is supported financially by National Natural Science Foundation of China (Grant Nos. 61127012, 61307122 and 61475043), University Science and Technology Innovation Team Support Project of Henan Province (Grant No. 13IRTTHN016), and Basic and Advanced Technology Research Program of Henan Province (Grant No. 142102210450).

Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.infrared.2015.10. 001. References [1] L. Del Campo, R.B. Pérez-Sáez, X. Esquisabel, I. Fernández, M.J. Tello, New experimental device for infrared spectral directional emissivity measurements in a controlled environment, Meas. Sci. Technol. 77 (2006) 113111. [2] Y.F. Liu, Z.L. Hu, D.H. Shi, K. Yu, Experimental investigation of emissivity of steel, Int. J. Thermophys. 34 (2013) 496–506. [3] P. Coppa, A. Consorti, Normal emissivity of samples surrounded by surfaces at diverse temperatures, Measurement 38 (2005) 124–131. [4] G. Cao, S.J. Weber, S.O. Martin, T.L. Malaney, S.R. Slattery, M.H. Anderson, K. Sridharan, T.R. Allen, In situ measurements of spectral emissivity of materials for very high temperature reactors, Nucl. Technol. 175 (2011) 460–467. [5] A. Seifter, F. Sachsenhofer, G. Pottlacher, A fast laser polarimeter improving a microsecond pulse heating system, Int. J. Thermophys. 23 (2002) 1267–1280. [6] G. Teodorescu, P.D. Jones, R.A. Overfelt, B. Guo, Spectral-directional emittance of 99.99% aluminum, thermally oxidized below its melting point, Int. J. Thermophys. 27 (2006) 554–568. [7] T. Furukawa, T. Iuchi, Experimental apparatus for radiometric emissivity measurements of metals, Rev. Sci. Instrum. 71 (2000) 2843–2847. [8] R. Brandt, C. Bird, G. Neuer, Emissivity reference paints for high temperature applications, Measurement 41 (2008) 731–736. [9] G. Pottlacher, A. Seifter, Microsecond laser polarimetry for emissivity measurements on liquid metals at high temperatures—application to tantalum, Int. J. Thermophys. 23 (2002) 1281–1291. [10] C. Li, F. Yi, J. Dai, Normal spectral emissivity changes of tungsten at 633 nm in the range from room temperature to melting point under pulse-heating conditions, Photon. Asia 2002 (2002) 305–308. [11] F. Wang, L. Cheng, H. Mei, Q. Zhang, L. Zhang, Effect of surface microstructures on the infrared emissivity of graphite, Int. J. Thermophys. 35 (2014) 62–75. [12] B. Hay, J. Hameury, N. Fleurence, P. Lacipiere, M. Grelard, V. Scoarnec, G. Davee, New facilities for the measurements of high-temperature thermophysical properties at LNE, Int. J. Thermophys. 35 (2013) 1712–1724. [13] B. Zhang, J. Redgrove, J. Clark, A transient method for total emissivity determination, Int. J. Thermophys. 25 (2004) 423–438. [14] V. Schmidt, S. Meitzner, H. Sandring, P. King, Automatic emissivity measurement setup for industrial radiation thermometry, TEMPERATURE: Its Measurement and Control in Science and Industry; Volume VII; Eighth Temperature Symposium, vol. 684, AIP Publishing, 2003, pp. 723–728. [15] S. Hatzl, M. Kirschner, V. Lippig, T. Sander, C. Mundt, M. Pfitzner, Direct measurements of infrared normal spectral emissivity of solid materials for high-temperature applications, Int. J. Thermophys. 34 (2013) 2089–2101. [16] G. Teodorescu, Radiative emissivity of metals and oxidized metals at high temperature, Ph.D. thesis, Auburn University, Alabama, USA, 2007. [17] X. Shen, L. Li, X. Wu, Z. Gao, G. Xu, Infrared emissivity of Sr doped lanthanum manganites in coating form, J. Alloy. Compd. 509 (2011) 8116–8119. [18] C.D. Wen, I. Mudawar, Emissivity characteristics of roughened aluminum alloy surfaces and assessment of multispectral radiation thermometry (MRT) emissivity models, Int. J. Heat Mass Trans. 47 (2004) 3591–3605. [19] M.F. Modest, Radiative Heat Transfer, Academic Press, 2013. [20] L.M. Hanssen, S.N. Mekhontsev, V.B. Khromchenko, Infrared spectral emissivity characterization facility at NIST, Defense and Security, April, International Society for Optics and Photonics, 2004, pp. 1–12. [21] G. Cao, S.J. Weber, S.O. Martin, K. Sridharan, M.H. Anderson, T.R. Allen, Spectral emissivity of candidate alloys for very high temperature reactors in high temperature air environment, J. Nucl. Mater. 441 (1) (2013) 667–673. [22] L. Del Campo, R.B. Pérez-Sáez, M.J. Tello, Iron oxidation kinetics study by using infrared spectral emissivity measurements below 570 °C, Corros. Sci. 50 (1) (2008) 194–199. [23] P.C. Dufour, N.L. Rowell, A.G. Steele, Fourier-transform radiation thermometry: measurements and uncertainties, Appl. Opt. 37 (1998) 5923–5931. [24] J. Ishii, A. Ono, Uncertainty estimation for emissivity measurements near room temperature with a Fourier transform spectrometer, Meas. Sci. Technol. 12 (2001) 2103. [25] L. Del Campo, R.B. Pérez-Sáez, L. González-Fernández, M.J. Tello, Combined standard uncertainty in direct emissivity measurements, J. Appl. Phys. 107 (2010) 113510. [26] G. Teodorescu, P.D. Jones, R.A. Overfelt, B. Guo, High temperature emissivity of high purity titanium and zirconium, in: Proceedings of the Sixteenth Symposium on Thermophysical Properties, 2006. [27] J. Ishii, M. Kobayashi, F. Sakuma, Effective emissivities of black-body cavities with grooved cylinders, Metrologia 35 (1998) 175–180.