Vanadium Dioxide as a Natural Disordered Metamaterial: Perfect Thermal Emission and Large Broadband Negative Differential Thermal Emittance

We experimentally demonstrate that a thin (approximately $150\mathrm{\text{−}}\mathrm{nm}$) film of vanadium dioxide (${\mathrm{VO}}_2$) deposited on sapphire has an anomalous thermal emittance profile when heated, which arises because of the optical interaction between the film and the substrate when the ${\mathrm{VO}}_2$ is at an intermediate state of its insulator-metal transition (IMT). Within the IMT region, the ${\mathrm{VO}}_2$ film comprises nanoscale islands of the metal and dielectric phases and can thus be viewed as a natural, disordered metamaterial. This structure displays “perfect” blackbodylike thermal emissivity over a narrow wavelength range (approximately $40{\mathrm{cm}}^{-1}$), surpassing the emissivity of our black-soot reference. We observe large broadband negative differential thermal emittance over a $>10°\mathrm{C}$ range: Upon heating, the ${\mathrm{VO}}_2\mathrm{\text{−}}\mathrm{\text{sapphire}}$ structure emits less thermal radiation and appears colder on an infrared camera. Our experimental approach allows for a direct measurement and extraction of wavelength- and temperature-dependent thermal emittance. We anticipate that emissivity engineering with thin-film geometries comprising ${\mathrm{VO}}_2$ and other thermochromic materials will find applications in infrared camouflage, thermal regulation, and infrared tagging and labeling.

This article appears in the following collection:

Popular Summary

An electric stovetop glowing red, a conventional light bulb producing a familiar warm glow, and even sunlight are all examples of thermal radiation—emission of light from any object at a temperature above absolute zero. The spectrum and intensity of this thermal radiation generally depend on the temperature as well as a factor called the emissivity, which is usually independent of the object’s temperature. For a conventional emitter of thermal radiation, the total power emitted is proportional to the fourth power of the temperature expressed in the kelvin scale. This is confirmed by our everyday experiences: The hotter an object is, the more it glows.

If the emissivity can be engineered to be temperature dependent, however, many opportunities arise: One example is “smart” thermal devices that keep heat in when cold and lose more when hot, or vice versa, depending on the desired application. In this work, we have designed a material with an emissivity value that varies widely with temperature by depositing a thin film of vanadium oxide ($\mathrm{V}{\mathrm{O}}_2$) on a sapphire substrate.

Vanadium oxide ($\mathrm{V}{\mathrm{O}}_2$) is a so-called phase-change material that undergoes a structural and electronic phase transition at approximately $70°\mathrm{C}$. During the course of this transition, the material shows a coexistence of metal and dielectric phases and, as a result, demonstrates highly tunable optical properties by temperature. Because of this phase coexistance, $\mathrm{V}{\mathrm{O}}_2$ near its phase transition can be thought of as a natural, disordered, tunable metamaterial. When our heated sample reaches about $65°\mathrm{C}$, the emissivity begins to rise, reaching a maximum at around $75°\mathrm{C}$, and then drops dramatically before settling at about $85°\mathrm{C}$. The effect is so large that our sample emits half of the thermal radiation at $100°\mathrm{C}$ that it does at $75°\mathrm{C}$—a remarkable contrast to the behavior of conventional thermal emitters. We now call the effect “negative differential thermal emittance.”

We envision that this type of unconventional thermal emittance can find numerous applications in infrared camouflage, thermal regulation, and infrared tagging and labeling. We also believe that this experiment will further encourage the exploration of the transition region of phase-change materials for optical and optoelectronic device applications.

Figure 1

Experimental setup. The ${\mathrm{VO}}_2\mathrm{\text{−}}\mathrm{\text{sapphire}}$ sample is mounted on a temperature-controlled stage, and the thermal emission is sent into an FTIR spectrometer equipped with a DTGS detector.

Figure 2

(a),(b) Experimentally determined evolution of the ${\mathrm{VO}}_2\mathrm{\text{−}}\mathrm{\text{sapphire}}$ emissivity for increasing temperature, separated into ranges of $35°\mathrm{C}–74.5°\mathrm{C}$ and $74.5°\mathrm{C}–100°\mathrm{C}$ for visual clarity. (c) Thermal emission density (spectral radiance) from black soot (dashed lines) and our ${\mathrm{VO}}_2\mathrm{\text{−}}\mathrm{\text{sapphire}}$ sample (solid lines) for three different temperatures. The data were taken for increasing temperatures. (d) Thermal emissivity of the ${\mathrm{VO}}_2\mathrm{\text{−}}\mathrm{\text{sapphire}}$ sample at a wave number of $864{\mathrm{cm}}^{-1}$ for heating (solid line) and cooling (dashed line), respectively. The dotted line denotes the assumed emissivity of the black-soot reference ($\epsilon =0.96$).

Figure 3

(a) Emitted power of the ${\mathrm{VO}}_2\mathrm{\text{−}}\mathrm{\text{sapphire}}$ sample integrated over the 8- to $14\mathrm{\text{−}}\mu \mathrm{m}$ atmospheric transmission window for heating (solid line) and cooling (dashed line), compared to the emitted power from black soot. (b) The integrated emissivity of the ${\mathrm{VO}}_2\mathrm{\text{−}}\mathrm{\text{sapphire}}$ sample over the 8- to $14\mathrm{\text{−}}\mu \mathrm{m}$ wavelength range. (c). Infrared camera images of the sample ($\mathrm{\text{diameter}}=1\mathrm{cm}$) for increasing temperatures.