Emissivity measurement splits into two broad families: direct methods, which capture the thermal radiation a surface actually emits and compare it against an ideal emitter, and indirect methods, which measure how much radiation a surface reflects and calculate emissivity from that. Each family has trade-offs in accuracy, temperature range, and practicality, and choosing the right technique depends on the material, the conditions, and how much control you have over the measurement environment. Complicating everything is the fact that emissivity is not a fixed number for a given material; it shifts with temperature, wavelength, surface roughness, and oxidation state, so the measurement method has to account for whatever is changing.
Direct Radiometric Measurement
The most intuitive way to measure emissivity is to point an infrared detector at a sample and measure how much thermal radiation it produces, then compare that reading against a reference blackbody at the same temperature. A blackbody, at least in theory, emits the maximum possible radiation at any given temperature, so the ratio of your sample’s radiation to the blackbody’s radiation gives you emissivity directly.1Cryogenics. Preliminary development of emissivity measurement system at low temperature based on radiometric method This approach is grounded in Kirchhoff’s law of thermal radiation, which states that at thermal equilibrium, a surface’s emissivity equals its absorptivity.2PubMed. Generalization of Kirchhoff’s law of thermal radiation: the inherent relations between quantum efficiency and emissivity
In practice, a direct radiometric setup needs several things working well at once. You need a reference blackbody source with a known, very high emissivity (typically above 0.99). You need the sample and the blackbody at the same temperature, which is straightforward at room temperature but increasingly tricky as you move to extremes. And you need a detector sensitive enough to distinguish meaningful differences in radiation intensity. One facility at the PROMES-CNRS laboratory, for example, measures total or spectral radiance from a sample and divides it by the corresponding blackbody radiance at the same temperature and wavelength range to get directional emissivity.3Ceramics International. Emissivity, catalycity and microstructural characterization of ZrB2–SiCfiber based UHTC at high temperature in a non-equilibrium air plasma flow
The strength of the direct radiometric approach is conceptual clarity: you measure what the surface actually emits, not an inferred quantity. The weakness is that any error in knowing the sample’s true temperature shows up directly in the emissivity value. If you are off by even a few degrees at high temperatures, the calculated emissivity shifts, and at lower temperatures where radiation signals are weak, noise becomes a serious problem.
The Calorimetric Approach
A second direct method sidesteps the detector-comparison problem entirely. Instead of measuring radiation with an optical sensor, a calorimetric system measures the total power a sample radiates by tracking the electrical energy needed to maintain it at a known temperature. A thin strip of the material under test is electrically heated inside a vacuum chamber; the vacuum eliminates convective heat loss, so the only way the sample sheds energy is by radiation. By measuring the electrical input power and accounting for conduction through the support wires, you can calculate total hemispherical emissivity.4Measurement Science and Technology. A steady-state measurement system for total hemispherical emissivity
This technique gives you total hemispherical emissivity, meaning the radiation emitted in all directions and across all wavelengths at once. That makes it especially useful for engineering applications like spacecraft thermal management or furnace design, where what you care about is total heat loss, not spectral detail. The downside is that it requires the sample to be conductive (since you are passing current through it) and thin enough to reach thermal equilibrium quickly. It also tells you nothing about how emissivity varies with wavelength.
Indirect Methods Through Reflectance
Indirect measurement flips the problem. Instead of detecting emitted radiation, you measure how much radiation a surface reflects and then calculate emissivity from the relationship between the two. For opaque materials (which do not transmit radiation), emissivity equals one minus reflectance. This is a direct consequence of energy conservation: radiation hitting an opaque surface either gets absorbed or reflected, and absorbed radiation equals emitted radiation at equilibrium.
The most common indirect setup pairs a Fourier Transform Infrared Spectrometer (FTIR) with a gold-coated integrating sphere. The sphere collects reflected light from all directions, giving you the total directional-hemispherical reflectance. One such system can handle both specular and diffuse samples at temperatures ranging from room temperature up to about 475 K.5International Journal of Thermophysics. Comparison of Direct and Indirect Methods of Spectral Infrared Emittance Measurement A more recent apparatus used a gold-coated diffuse integrating sphere placed directly in a spectrometer’s sample chamber, with a dual-sided gold-coated mirror capturing both the incident and reflected light paths.6Infrared Physics & Technology. Normal spectral emissivity measurement of thermal management materials for electronic components in the 2–14 μm range
The appeal of indirect methods is that they work well near room temperature, where emitted radiation is weak and hard to detect directly. They also give you spectral resolution across a broad wavelength range, which is critical for applications like radiative cooling, where you need to know how a surface behaves in specific infrared windows. Characterization of passive radiative cooling materials, for instance, requires emissivity data spanning roughly 250 nm to 50 µm with absolute uncertainties below 0.03.7International Journal of Thermophysics. Emissivity and Reflectivity Measurements for Passive Radiative Cooling Technologies
The limitation is the assumption of opacity. If the material transmits any radiation, the simple “one minus reflectance” formula breaks down, and you get a number that overestimates emissivity. You also need a very good reference mirror or reference coating to calibrate the sphere, and any degradation of the gold coating over time introduces systematic error.
Why Surface Condition Changes Everything
One of the most practically important things about emissivity measurement is that two samples of the “same” material can give wildly different readings if their surfaces differ. Roughness, oxidation, coatings, and even handling can shift emissivity enough to invalidate a measurement if you are not paying attention.
Surface roughness increases emissivity because a rough surface has more effective area to interact with radiation. The wrinkles and valleys trap and absorb incoming photons more efficiently than a smooth surface does.8Light: Science & Applications. A straightforward spectral emissivity estimating method based on constructing random rough surfaces The magnitude of this effect is not trivial: one study on heat-sink surfaces found that increasing roughness could raise thermal emissivity by up to 2.5 times.9PubMed Central. Effect of Surface Microstructure on the Heat Dissipation Performance of Heat Sinks Used in Electronic Devices
Oxidation is an even more dramatic factor for metals. A 316L stainless-steel alloy tested at 1273 K had an emissivity of just 0.055 in its unoxidized state, but after 50 hours of thermal oxidation that value climbed to 0.829.10Infrared Physics & Technology. Normal spectral emissivity measurement study of high-temperature alloy (316L) at 673–1273 K with a thermal oxidation surface That is a fifteenfold increase, purely from the oxide layer that formed during heating. The relationship between oxide thickness and emissivity is not linear, either. A study of TA15 titanium alloy found that below an oxide thickness of about 0.3 µm, emissivity stayed nearly constant around 0.18, but once the layer grew past that threshold, emissivity rose rapidly, reaching 0.67 at a thickness of 38 µm.11Journal of Alloys and Compounds. Oxidation behavior of TA15 titanium alloy at high-temperature and the effect on infrared emissivity
The practical takeaway is that any emissivity measurement of a metal at high temperature needs to document the surface condition carefully. An emissivity value measured on a freshly polished sample in a vacuum may be nearly useless for predicting the behavior of the same alloy after it has been in service and developed an oxide layer. This is also why lookup tables of “emissivity of stainless steel” can be misleading; the value depends heavily on whether the surface is polished, brushed, oxidized, or painted.
Semi-Transparent Materials Need a Different Approach
Opaque materials keep things relatively simple because radiation interacts only at or very near the surface. Semi-transparent materials, like certain ceramics, glasses, and thin polymer films, are a different story. These materials not only reflect and emit radiation but also transmit it, and both emission and scattering happen throughout the material’s volume rather than just at the surface.12Journal of Quantitative Spectroscopy and Radiative Transfer. Apparent emissivity measurement of semi-transparent materials part 1: Experimental realization
This means that a standard reflectance-based measurement will give you an incorrect emissivity because it does not account for the transmitted fraction. And a direct radiometric measurement captures what is called “apparent emissivity,” which includes radiation emitted from inside the material at temperatures that may differ from the surface. To get a meaningful number, you need to know the temperature distribution and spectral radiance throughout the entire observed volume, not just at the surface. In practice, this often means combining measurements of reflectance, transmittance, and surface temperature simultaneously, then working backward through a radiative-transfer model. It is significantly more involved than measuring an opaque metal, and the uncertainty is correspondingly larger.
Multispectral Pyrometry and Real-World Temperature Measurement
In many industrial settings, you cannot bring the sample into a laboratory. You need to measure the temperature of a hot surface in situ, and to do that accurately with a radiation thermometer, you need to know its emissivity. The catch is circular: to measure temperature you need emissivity, and to measure emissivity you need temperature. Multispectral pyrometry tries to break this circularity by measuring thermal radiation at many wavelengths simultaneously and solving for both temperature and emissivity together.
A recent development uses a nine-channel multispectral imaging system that captures radiative energy across multiple narrow-band wavelengths at the same time. By fitting the resulting spectral data to emissivity models that account for both temperature and wavelength dependence, the system can produce full-field temperature maps without needing a pre-measured emissivity value.13Optics and Lasers in Engineering. Improved multispectral pyrometry for synchronous full-field temperature and deformation measurement at elevated temperatures This is particularly valuable in scenarios like metal forming or additive manufacturing, where the surface is evolving during the measurement and no single emissivity value applies.
The limitation of multispectral pyrometry is that it relies on assumptions about how emissivity varies with wavelength. If the real surface does not follow the assumed model, the solved temperature can be significantly off. More wavelength channels reduce this risk but add optical complexity and cost.
Fast Transient Methods for Extreme Temperatures
Measuring emissivity at temperatures above about 1500 K gets progressively harder because the sample starts changing. Oxidation accelerates, grain structures evolve, and some materials begin to sublimate. Holding a sample at these temperatures long enough for a steady-state measurement can alter the very surface you are trying to characterize.
Fast transient (or “dynamic”) methods address this by heating the sample for only a very short time and capturing the emissivity data before the surface has time to change. One such system uses a modified laser-flash apparatus: a high-energy laser pulse heats the specimen, and a fast-response radiation thermometer records the temperature rise at the same wavelength as the laser. By knowing the laser pulse energy and the resulting temperature jump, the normal spectral emissivity can be calculated. This method has been demonstrated on isostatic-pressed graphite between 1000 K and 1800 K, with corrections applied for the surrounding furnace radiation and for the fact that the temperature rise is not perfectly adiabatic.14International Journal of Thermophysics. A Dynamic Method to Measure Emissivity at High Temperatures
The advantage is clear: the measurement happens in milliseconds, so the surface does not have time to oxidize or degrade during the pulse. The disadvantage is that you are measuring emissivity at a single wavelength (typically the laser wavelength), so you do not get the full spectral picture.
Measurement at Temperature Extremes
Emissivity measurement is not confined to comfortable laboratory temperatures. Space applications in particular demand data at cryogenic temperatures where radiation signals are vanishingly small. A system designed for this purpose uses a Gifford-McMahon cryocooler to bring the sample plate down to about 9 K, with surrounding thermal shields cooled to roughly 47 K to minimize background radiation. Using an HgCdTe detector sensitive in the 2–12 µm range, the system has measured the emissivity of materials including Nextel 811-21 coating, 304 stainless steel, and G10 composite in the range of 240 K to 300 K.1Cryogenics. Preliminary development of emissivity measurement system at low temperature based on radiometric method
Getting reliable readings at these temperatures is hard because the radiation emitted by a cold surface is extremely weak, and any stray thermal radiation from the chamber walls, the detector housing, or even the optical path can overwhelm the signal. The cold shields and vacuum environment are essential to keep the noise floor below the signal you are trying to measure.
At the other extreme, facilities like the University of the Basque Country have upgraded their direct emissivity measurement systems to cover 300 K to 1273 K under high vacuum, with refined measurement equations and ISO-compliant uncertainty budgets. Their upgrades include Monte Carlo procedures for converting spectral emissivity data into accurate total emissivity values.15Metrologia. Updated measurement method and uncertainty budget for direct emissivity measurements at the University of the Basque Country This kind of metrological rigor matters because, at high temperatures, even small uncertainties in spectral emissivity can compound into significant errors in total emissivity when integrated across all wavelengths.
Portable Instruments and Field Measurements
Not every emissivity measurement happens in a vacuum chamber. Building inspectors use infrared cameras to find heat leaks. Solar energy engineers need to verify that receiver coatings meet their specifications on installed hardware. For these applications, portable emissometers and reflectometers offer a way to get a reading on-site, without removing a sample.
One such instrument, the SOC 410-VIS-IR modular solar reflectometer/emissometer, combines two measurement heads: one for solar reflectance and one for thermal emittance. It can be transported into the field and used on both flat and curved surfaces.16Solar Energy Materials and Solar Cells. Laboratory intercomparison of solar absorptance and thermal emittance measurements at room temperature The trade-off with portable devices is always accuracy versus convenience. A laboratory integrating sphere and a calibrated blackbody will outperform a handheld device, but the handheld device can go places the laboratory cannot.
A relevant finding from building-materials research highlights why field accuracy matters. When researchers compared emissivity values obtained using an emissometer and an infrared camera with a black-tape reference method against published literature values, they found meaningful disagreements. Those disagreements translated to surface temperature differences of up to 7 °C.17PubMed Central. Emissivity of Building Materials for Infrared Measurements For a building energy audit, a 7 °C error could mean diagnosing a thermal bridge that does not exist, or missing one that does.
How Space Environments Degrade Emissivity Over Time
Emissivity is not just hard to measure; it changes over time as materials are exposed to their operating environment. Spacecraft thermal control surfaces are a stark example. A study simulating three years of geostationary orbit exposure, including ultraviolet radiation and charged-particle bombardment, measured the reflectance of various thermal control materials in situ. Among low-solar-absorptance materials, white paints showed the most significant degradation, while rigid optical solar reflectors remained relatively stable.18Advances in Space Research. Degradation of thermal control materials under a simulated radiative space environment
This kind of degradation is why beginning-of-life emissivity values are not sufficient for spacecraft design. Engineers need to know how emissivity and absorptance will shift over the mission lifetime, which means measuring not only the fresh material but also samples that have undergone accelerated environmental aging. The same principle applies in terrestrial applications: a furnace liner measured when new may have very different emissivity after months of thermal cycling, and a rooftop cooling coating measured in the lab may not match the performance of the same coating after a year of UV exposure and particulate contamination.
This time-dependent behavior circles back to the fundamental challenge of emissivity measurement. A single number from a single test, under controlled conditions, on a pristine sample, tells you what that specific surface did at that specific moment. Translating that into a prediction of real-world thermal performance requires understanding how the surface will evolve and choosing a measurement method that captures the conditions closest to the ones the material will actually face.