Photoinduced photon avalanche turns white objects into bright blackbodies

In fact, an absolute blackbody is only a physical abstraction that does not exist in nature. A small hole on a closed cavity is generally considered to approximate an ideal blackbody. When light enters the hole from the outside, it has only a small chance of escaping through the hole by reflection from the inner wall¹. In the real world, however, few objects have the absorptive properties of the ideal blackbody—strong absorption at all wavelengths. The optical absorption properties of most materials are intermediate between those of whitebody and blackbody, and the possibility of interconversion between whitebody and blackbody has never been studied.

In 2002, our research group first reported the wideband upconversion luminescence of TiO₂:Mo excited by near-infrared light (NIR), and confirmed that the upconversion luminescence had the characteristics of photon avalanche². Since 2010, similar phenomena have been reported by Peter^3,4,5, Strek^{6,7,8,9,10,11,12,13}, Song^14,15,16,17, Liu^18,19,20,21, Yan²², and other research groups^{23,24,25,26,27,28}. The researchers observed wideband upconversion luminescence from NIR to visible light using a variety of different samples²¹. While researchers have provided different explanations for the physical processes occurring in these samples^{2,4,6,16,20,22,29,30}, the actual physical mechanism has remained a mystery to this day, and exploring the physics involved remains a meaningful challenge.

Many experimental results demonstrated that, under intense laser irradiation, samples emitted bright incandescent light, and the color temperatures obtained from fitting spectra were considerably higher than the samples or sample holders could bear. The phenomenon also occurred at low temperatures (~10 K)^4,6,30,31,32, which confused researchers. Further investigation showed that the wideband electromagnetic waves emitted by the samples were very similar to the blackbody radiation and contained not only upconversion luminescence but also down-conversion (or down-shift) luminescence^33,34,35. To be precise, after the photon-avalanche optical frequency conversion, the samples converted the irradiated light into wideband electromagnetic radiation with a spectrum very similar to that of blackbody radiation (or thermal radiation). Therefore, we call it here photoinduced blackbody radiation (PBR) or photon avalanche blackbody radiation (PABR).

To verify whether the change in optical absorption is related to the temperature of a sample, we measured the optical absorption of samples when they were heated. The experimental results show that electric or flame heating does not have a noticeable effect on their absorption. It is clear that the irradiated light field with a high-power density is the fundamental factor in changing the absorption properties of samples. Here we report our latest results: white powder samples transform rapidly into blackbodies under intense light and return to their original states (e.g. white powder) when light irradiation is removed. Under the focused NIR laser irradiation, the samples (usually white color) emitted intense broadband light radiation, and the optical absorption in the ultraviolet, visible and infrared regions increased significantly and rapidly, by >90%, like blackbodies. In other words, when these samples emitted intense incandescent light, their optical absorption suddenly increased in an avalanche way. That is why we call it “bright blackbody”.

Experimentally, we confirm that this phenomenon has four characteristics: photon avalanche optical frequency conversion, broadband light radiation, optical bistable luminescence, and full-band strong absorption. When the recorded spectra are fitted using Planck’s law, we note that the fitting results are significantly different with and without light irradiation. Dynamic analysis showed that, when the irradiated light was removed for a period, the sample indeed emitted thermal radiation satisfying Planck’s formula. However, in the presence of irradiated light, the broadband conversion luminescence deviated from the Planck’s curve and the actual temperature detected by a platinum-rhodium thermocouple was considerably lower than the color temperature. Photoconductivity measurements indicated that photoinduced blackbodies were accompanied by the formation of free carriers or charged particles (free electrons and/or free ions) in samples. This result suggests that the photoinduced charged particles is closely related to the new energy states generated in the PBR, resulting in a sharp increase of optical absorption in full spectral range.

Broadband absorption of photoinduced blackbodies

The PBR phenomenon reported here generated from the irradiation of focused intense laser. In other words, the broadband electromagnetic radiation we produced in our experiments was not by heating, but by lighting. The materials that became bright blackbodies in the experiments were not black at all in the absence of pump laser. Although the broadband conversion luminescence has been observed in a variety of materials, two representative white powder materials, Y₂O₃ and Yb₂O₃, are used as examples here in order to clearly illustrate the issue. Their absorptions at 980 nm are completely different, with almost no absorption by Y₂O₃, whereas the Yb³⁺ ions in Yb₂O₃ have a strong absorption.

Although Y₂O₃ powder sample hardly absorb 980 nm NIR, the PBR phenomenon will still occur when the focusing 980 nm laser is strong enough. Figure 1a shows the variation in the intensities of luminescence and scattered 980 nm laser from Y₂O₃ powder before and after the occurrence of PBR phenomenon. When the laser power increased from 0 W to 22 W, Y₂O₃ did not emit observable visible light. When the pump power exceeded 22 W, the sample suddenly emitted intense white light (See Supplementary Methods and Supplementary Fig. 1 in detail) with a broadband spectrum. At the same time, the intensity of scattered laser from the Y₂O₃ powder suddenly decreased greatly, as shown in Fig. 1a and Supplementary Fig. 2, indicating that the broad-spectrum emission was accompanied by a sudden sharp increase in the absorption of pump laser by the sample. In the above process, we can see that the sudden generation of broadband luminescence and the rapid increase of optical absorption occur synchronously after an apparent energy accumulation process. These features are similar to photon avalanche upconversion luminescence. The reported avalanche upconversion luminescence phenomena generally satisfy the following conditions: weak ground state absorption (GSA), strong excited state absorption (ESA), and an effective cross-relaxation (CR) between two transitions of \(|{E}_{2}\rangle \to |{E}_{1}\rangle\) and \(|{E}_{0}\rangle \to |{E}_{1}\rangle\) to populate the excited state \(|{E}_{1}\rangle\) gradually, as shown in Fig. 1b. When the population of the excited state \(|{E}_{1}\rangle\) reaches a certain level, the excited state absorption and cross-relaxation increase rapidly, leading to a sharp increase in the radiative transition from the upper level \(|{E}_{2}\rangle\) in an avalanche way. The enhancement of ESA and the increase of the excited state population constitute a positive feedback process, which is manifested by the sudden enhancement of upconversion luminescence. For rare earth ions, however, the enhanced absorption accompanying the photon avalanche comes from the ESA of lanthanide ions^{36,37,38,39,40,41,42}. But in fact, the materials used in our experiments, such as Y₂O₃ and Yb₂O₃, do not have such an excited state before PBR appears. Moreover, when PBR phenomenon occurs, any material used in the experiments will strongly absorb the irradiated light, regardless of the wavelength. These two features strongly suggest that these energy levels with broadband strong absorption are quantum states arose from the photoinduction under strong light irradiation, which are different from the original ones. As such, the photon avalanche luminescence observed in the present case should have a different physical mechanism. Hereafter, we specifically refer to the energy states generated in PBR as PBR quantum states or PBR levels, and to the specific avalanche luminescence exclusively due to PBR as PBR avalanche luminescence. The issue about their generations will be discussed and confirmed in more detail later with the experimental data from photoconductivity measurements.

a The changed intensities of luminescence (400 nm–900 nm) and scattered 980 nm laser (960 nm–980 nm) from Yb₂O₃ powder before and after the occurrence of PBR phenomenon. b Schematic of photon avalanche upconversion luminescence. c Variations of the scattering intensities of 5 probe lasers (266 nm, 405 nm, 532 nm, 650 nm, 808 nm and 1560 nm) before (red) and after (blue) the occurrence of PBR phenomenon. d In the PBR state, the sample’s relative absorption (absorption ratios) to the six probe lasers are above 90%, showing blackbody absorption characteristics. The relative absorption at 980 nm (red dot) was measured upon 808 nm laser irradiation, as shown in Supplementary Fig. 3.

Full size image

To probe the wavelength range of the enhanced absorption, we examined the intensities of the scattered lasers with different wavelengths. We simultaneously irradiated one pump laser (980 nm, 30 W) and six low-power (~100 mW) probe lasers (266 nm, 405 nm, 532 nm, 650 nm, 808 nm, and 1560 nm) onto Yb₂O₃ powder. When the sample turn into photoinduced blackbody under the irradiation of the pump laser, the scattered intensities of six probe lasers suddenly decreased greatly, as shown in Fig. 1c and Supplementary Fig. 3, indicating that the sample produced strong optical absorption in a large spectral range. As shown in Fig. 1d, the absorption ratios for the six probe lasers are all above 90%, exhibiting broadband absorption characteristics similar to that of a blackbody. Therefore, we refer to this phenomenon as photoinduced blackbody radiation and also name the blackbody produced in PBR photoinduced blackbody. In addition, to study the absorption characteristics of photoinduced blackbodies in more detail, we measured the absorption ratios of an Yb₂O₃ sheet sample in the PBR state using a lock-in amplifier and a wide-band light source. The schematic diagram of the optical transmission (absorption) experiments and measurement results are shown in Supplementary Fig. 4 and Supplementary Fig. 5, respectively. It can be seen from Supplementary Fig. 5 that the thin Yb₂O₃ sheet in PBR state has a strong optical absorption in the range of 500 nm–1500 nm. With the increase of 980 nm pump laser power, the optical transmission also decreased continuously in the spectral range. The change of transmitted light intensity is mainly due to two factors: 1. Broadband absorption produced in the PBR state; 2. The enlarged PBR spot on the sample. The spot size of the detection light focused on the sample is fixed and larger than the PBR area, so the PBR spot cannot completely cover it. With the increase of irradiation power, the PBR region increases gradually, resulting in a larger area of strong absorption on the sample.

Factors affecting photoinduced blackbodies

In order to investigate the influencing factors on the PBR, we studied the relationship between the intensity of PBR and the pump power. The laser spot area we used was about 1 mm,² and the power range was 0 ~ 30 W. Therefore, the pump power density in this experiment was 0 ~ 30 W /mm.² As shown in Fig. 2a, with the increase and decrease of the pump laser power at 980 nm, the photon avalanche luminescence of Y₂O₃ powder has optical bistable characteristics. In the process of increasing the pump power from 0 W to 22 W, the sample did not emit light. As the pump power continued to increase to 22 W, the sample instantly entered PBR state and emitted strong incandescent light with a broad spectrum (See Supplementary Note 1 and Supplementary Fig. 12a in detail). At this moment, as shown in Supplementary Fig. 12e, the slope n = logI/logP increased to 70.5 instantly, showing the characteristics of photon avalanche luminescence. Therefore, 22 W was the power threshold for Y₂O₃ powder to enter the PBR state. As the pump power continued to increase to 30 W, the luminescence intensity continued to increase, and n remained at about 1.6 during the pump power continued to increase from 22 to 30 W. In the process of reducing the pump power from 30 W to 7 W, the sample kept in the PBR state although the broadband luminescence dimmed continuously. When the pump power was below 7 W, the sample quickly returns to its original white color and no longer emitted incandescent light. However, when the irradiated laser power was kept at 7 W, the photoinduced blackbody was stable in the state of continuous radiation. Therefore, in this case, the minimum power required to maintain PBR was 7 W. During the rise and fall of pump power, the luminescence intensity and the scattered 980 nm laser took two different paths, respectively, to form a bistable luminescence, as shown in Fig. 2a, b. Then, a small amount (~ 0.7 mol %) of Yb₂O₃ was mixed in the Y₂O₃ powder to increase its optical absorption at 980 nm. In this case, we still observed the bistable characteristics. However, as shown in Fig. 2b, the power threshold of the mixed sample was only 13 W, much lower than that of pure Y₂O₃, while the self-sustaining power still kept at 7 W, basically unchanged. Corresponding to the upper loops, similar bistable phenomena also can be found in the scattered laser, as shown in the lower part of Fig. 2a, b. When entering the PBR state, the optical absorption of the sample increased dramatically, while the scattering of pump laser decreased rapidly. The above experimental data indicate that the threshold power of PBR occurrence for a material is closely related to its initial absorption coefficient to the pump light. The larger the initial absorption coefficient, the lower the threshold, suggesting that the material with a strong optical absorption is more likely to enter the PBR state. On the other hand, there is no significant correlation between the minimum power density required to maintain the PBR and the initial absorption coefficient of the material.

a The PBR luminescence of Y₂O₃ powder has characteristics of photon avalanche and optical bistable luminescence. The occurrence of PBR coincides with a sudden drop in the intensity of the irradiation laser scattered by the sample. The red and black lines show the process of increasing and decreasing pump power, respectively. b The doping of 0.7 mol% Yb³⁺ increased the initial absorption of Y₂O₃ powder to 980 nm laser and reduced the threshold power of PBR occurrence, resulting in a shrunk bistable ring. c When the sample was heated to 1500 K, the optical absorption of the sample to the 808 nm probe laser (100 mW) remained essentially unchanged, but the threshold power of the sample entering the PBR state was effectively reduced to 5 W. d (I) 5 W 808 nm laser irradiation alone is not enough to make Yb₂O₃ powder enter PBR state; (II) The irradiation of 980 nm pulsed laser produced optical absorption at 808 nm in the Y₂O₃ powder, which resulted in the occurrence of PBR under the excitation of 5 W 808 nm laser; (III) Continued to stably remain in the PBR state after removing the 980 nm pulsed laser; (IV) When the 808 nm laser was blocked instantaneously by a mask, the PBR disappeared immediately.

Full size image

By comparing the upper and lower parts of Fig. 2a, b, it can be found that there are differences between the emission and the laser scattering bistable rings. In the process of reducing pump laser power, the PBR intensity of Y₂O₃ powder gradually decreased until it reached zero. Meanwhile, the scattering intensity of Y₂O₃ powder to pump laser was always at a very low level. However, when the sample exited the PBR state, there was a step increase in the intensity of the scattered laser, indicating that the absorption transition in the sample abruptly disappeared. The same was true for Y₂O₃:0.7 mol%Yb³⁺ sample, except that this sudden change in absorption was relatively more pronounced. The rapid loss of optical absorption accompanying the sample’s exit from the PBR state implies the sudden disappearance of the energy levels involved in the broadband absorption in the sample. Due to the intensity of the PBR is related to the temperature of the sample, and the response rate of the thermal effect is slow, the sudden disappearance of the optical absorption in the sample has not been reflected in the PBR bistable rings. It is well known that thermal bistable curves change slowly, but the four bistable curves in Fig. 2 all have sudden jumping processes during their state transitions, so it can be inferred that they are not caused by thermal effects. The generation and extinction of PBR quantum states in the samples should be the physical origin of the non-thermal optical bistable state.

To find out the similarities and differences between PBR and heat-induced blackbody radiation, and whether heating would affect the occurrence of PBR, we smeared Y₂O₃ powder on the surface of a Ni-Cr heater and irradiated the Y₂O₃ powder with an 808 nm laser, the test device is shown in Supplementary Fig. 6. While changing the 808 nm laser power, we monitored the 808 nm pump laser scattered from the Y₂O₃ powder with a spectrometer (Ocean Insight, HR4000CG-UV-NIR). When the power of 808 nm pump laser was 100 mW and the heater was not working, the intensity of detected scattered laser did not change with time. When Y₂O₃ powder was heated to 1500 K by the heater, the intensity of scattered laser fluctuated sharply with time, but its average value remained the same significantly, as shown in Fig. 2c. The sharp fluctuation should come from the sample vibration caused by heating. When the heating current was turned off, the temperature gradually returned to the room temperature, and the intensity of the detected scattered light gradually returned to its original stable level. The results show that, in the range of room temperature to 1500 K, heating does not affect the absorbance of Y₂O₃ at 808 nm or lead to the formation of PBR quantum states.

Subsequently, we increased the pump power of 808 nm laser to 5 W (below the threshold power), the PBR did not occurred and the scattered laser intensity remained stable, as shown in Fig. 2c. When the heating current was turned on, Y₂O₃ sample was heated to 1500 K and entered the PBR state within a few seconds. Currently, the intensity of scattered laser decreased by ~79.5% within 4 s. When the heating current was turned off, the sample continued to emit incandescent light for a period (~10 s), then suddenly exited from the PBR state and the scattered laser returned to its original intensity. These results indicate that the temperature of sample affect the difficulty of the PBR occurrence, and heating can effectively reduce the power threshold.

Then, Yb₂O₃ powder was irradiated with a 5 W 808 nm continuous laser and a 980 nm pulsed (40 ms, 10 Hz) laser with a peak power of 10 W, and the scattered light was monitored at 808 nm and 980 nm, respectively. The optical absorption of Yb₂O₃ is very low at 808 nm but high at 980 nm. However, the irradiation of either laser beam alone was not sufficient to induce Yb₂O₃ powder into PBR state, as shown in Fig. 2d. When the Yb₂O₃ sample was irradiated with a 5 W continuous 808 nm laser alone, the scattering laser intensity at 808 nm remained stable. When the 980 nm pulsed laser and 808 nm continuous laser were focused on the sample simultaneously, the scattered intensity of 808 nm laser decreased slightly, indicating that the irradiation of 980 nm pulsed laser improved the optical absorption at 808 nm. After keeping in this state for ~12 s, the sample entered the PBR state with the emission of intense incandescent light, and the scattered intensity at 808 nm decreased rapidly at the same time, shown in Fig. 2d. When the 980 nm pulsed laser was turned off, the scattered intensity of 808 nm laser increased slightly, but the sample was still in the PBR state, indicating that the irradiation of a pulsed laser at 980 nm did affect the optical absorption characteristics of the sample at 808 nm, which was observable before and after the occurrence of PBR. In other words, the irradiation of 980 nm pulsed laser led to an additional absorption transition at 808 nm that would not otherwise have existed. This absorption transition did not start from the ground state, but from the excited state that has just been created. When the 808 nm pump light was momentarily blocked by a mask, the PBR disappeared immediately, and the intensity of the scattered 808 nm laser returned to its original level, as shown in Fig. 2d, indicating that continuous irradiation was necessary to maintain the existence of PBR quantum states, and transient interruption of excitation light led to the disappearance of these PBR quantum states.

Spectral deviation, temperature deviation, and the formation mechanism of PBR

What is the spectral difference between PBR and thermal radiation (blackbody radiation) in the usual sense? In general, the spectrum of thermal radiation satisfies Planck’s blackbody radiation law^43,44:

where, λ represents the wavelength, h is the Planck constant, c is the speed of light, k is the Boltzmann constant, and T is the absolute temperature.

In order to accurately record the PBR spectrum, we used two standard halogen lamps (calibrated by Beijing Metrology and Testing Institute) with different color temperatures as the standard light source to calibrate the whole measurement system. The measuring system schematic diagram and instrument calibration data are shown in Supplementary Fig. 7 and Supplementary Figs. 13–17. The PBR spectra of Y₂O₃ pumped by 980 nm laser with different power and the fitted curves of Planck’s law are shown in Fig. 3a. When the pump laser power was 30 W, the PBR spectrum was in good agreement with the Planck’s law. When the pump power was gradually reduced, the spectra of PBR gradually deviated from the description of Planck’s law. The lower the power, the more obvious the deviation. To account for the deviation, we propose a possible mechanism for populating the correlated energy states in the sample, as shown in Fig. 3c. It is well known that thermal radiation generates from the radiative transitions between two quantum states inside an object. When the temperature is high enough, charged particles in the object will undergo intense thermal vibration and radiate wideband electromagnetic waves (including visible light), thus exhibiting the observable blackbody radiation characteristics. In the case of thermal equilibrium, the population of all energy states follows a Boltzmann distribution. In other words, Planck’s law at thermal equilibrium reflects the Boltzmann distribution on the spectrum. However, in the PBR phenomenon, there are two population mechanisms simultaneously: (1) the lighting population caused by the optical absorption transitions under intense light irradiation and (2) the thermal population which tends to satisfy the Boltzmann distribution. When the pump power is low, the thermal vibration between cations and anions is at a relative low level, the photoinduced absorption transitions produce higher population proportions in the higher energy states, the thermal equilibrium between energy states is destroyed, and the population proportion of each energy state deviates from the Boltzmann distribution, and therefore the PBR spectrum deviates from the description of Planck’s law. At this time, the sample temperature fitted by luminescence spectrum is higher than the actual temperature measured by a thermocouple. Under high power light irradiation, however, a large number of photons convert their energy into the thermal vibration of atoms or ions, and the sample’s temperature increases rapidly. The intense thermal vibrations cause the population of the higher energy states to relax rapidly to the lower energy states, the thermal population based on the Boltzmann distribution becomes dominant again in a very short time, the deviation caused by the lighting population is masked, and the PBR spectrum basically conforms to the description of Planck’s law. However, the actual temperature of the sample should be lower than that obtained by fitting the spectrum due to the populations of many energy states still comes from optical absorption transitions. Our experimental results confirmed the above judgment. As shown in Fig. 3b, each actual temperature measured by the Pt/Rh thermocouple is lower than the color temperature obtained by fitting the corresponding spectrum. In contrast, the spectral temperature of the laser-heated Pt/Rh thermocouple can match the actual temperature very well. The method of measuring the actual temperatures is shown in Supplementary Figs. 8, 9, and the detail experimental data are shown in Supplementary Note 2 and Supplementary Figs. 18–20. In addition, when the PBR occurred, both of color temperature or actual measured temperature were lower than the melting point of the samples.

a PBR spectra of Y₂O₃ pumped by 980 nm lasers with different powers are compared with the fitted curves of Planck’s law. b When the PBR occurred under the irradiation of 980 nm laser, the spectral fitting temperature of Yb₂O₃ powder was very different from the actual temperature measured by using a thermocouple. The red and blue spheres represent spectral temperature and actual temperature, respectively. c Mechanism diagram of PBR photon avalanche and PBR bistable luminescence.

Full size image

Time-resolved spectra of photoinduced blackbodies

To further verify the reasonability of the above viewpoints, we analyzed the dynamic process of PBR occurrence and evolution. The effect of heat accumulation on PBR can be effectively reduced when the sample is excited by a pulsed laser. As shown in Fig. 4a and Supplementary Fig. 10, Yb₂O₃ was irradiated by a 980 nm laser with different pulse widths, and the PBR intensity at 620 nm was monitored as a function of time. Due to the strong absorption at 980 nm, the Yb₂O₃ sample can rapidly accumulate energy to enter the PBR state when the pulsed laser is focused on it. The time-resolved spectrum exhibited that the occurrence of PBR was not instantaneous but took 1 ~ 2 microseconds to accumulate energy. When the pulse width was 5 ms, the PBR intensity did not reach its maximum value. With the increase of pulse width, the time of optical excitation prolonged, and the PBR intensity increased rapidly until to a stable level in a very short time. When the pulse width reached ~20 ms, the PBR intensity reached its maximum. Increasing the pulse width further did not increase the PBR intensity. The time-resolved spectra recorded at four time points (as marked in Fig. 4a) were compared with the fitted curves of Planck’s law. As shown in Fig. 4b, the first two time-resolved spectra recorded during the laser pulse cannot match Planck’s law well in the range from 400 nm to 600 nm. The last two time-resolved spectra were recorded after the laser pulse. In the absence of laser irradiation, the populations of energy states in the sample were redistributed to satisfy the Boltzmann distribution. In this case, the time-resolved spectra rapidly tended to the Planck’s curves.

a The PBR intensity of Yb₂O₃ powder at 620 nm evolved with the width of 980 nm pump laser pulse. b Time-resolved spectra recorded at four time points (shown in Fig. 4a) were compared with the corresponding curves fitted using Plank’s formula, and there were clear differences between them during the presence of pulsed light. c After the occurrence of PBR, its integrated intensity (400 ~ 800 nm) and the electrical conductivity of Yb₂O₃ powder increased with the increase of 980 nm laser pump power.

Full size image

Conductivity of photoinduced blackbodies

To prove that the generation of PBR quantum states is the core factor for the occurrence of PBR, we studied the conductivity changes of Yb₂O₃ powder pumped by 980 nm pulsed laser at different powers (The test system is shown in Supplementary Fig. 11). As shown in Fig. 4c, when the pump power increased from 1 W to 3 W, the photoconductivity of Yb₂O₃ decreased significantly instead of increasing. The significant decrease of the conductivity surely came from the increased temperature of the sample induced by the irradiation of the pump laser. When the pump power reached to 4 W, the sample entered the PBR state instantaneously, and its electrical conductivity also rose briefly. Then, due to the photothermal effect, the further increased temperature caused the sample lattice to vibrate more violently and the conductivity to decrease further. When the pump pulse ended, the conductivity dropped sharply. At this time, the sample exited from the PBR state, and the high resistance of the sample quickly revealed itself at high temperature. When the pump power increased further, the PBR intensity continued to increase, and the electrical conductivity also increased synchronously. When the pump power was increased to 7 W, the photoinduced conductance was already greater than the resistance caused by thermal vibration, and the conductivity of the sample increased by 2.45 times. The conductivity maintained at a very high level during the duration of the light pulse. When the pump pulse ended, the electrical conductivity decreased sharply, and the resistance caused by accumulated heat rapidly appeared. In general, heating leads to an increase in the resistance of a material, while in PBR state, intense light irradiation leads to an increase in the conductivity of the material, which means that the intense light irradiation produces free charge carriers in the material, accompanied by strong absorption over a wide spectral range. This result suggests that the generation of photoinduced free charge carriers is closely related to the generation of PBR quantum states with the broadband strong absorption.

By irradiation with a focused laser, we changed white powder samples into excellent optical absorbers with broadband optical absorption characteristics, and at the same time, the samples emitted broadband electromagnetic waves like the radiation from a hot blackbody. We call this optical phenomenon photoinduced blackbody radiation (PBR). The blackbody produced by laser irradiation is bright, but it has a very high absorption rate for any wavelength of light. During the transformation from a white matter to a photoinduced blackbody, the broadband optical radiation showed the feather of photon avalanche luminescence. The reason for the photon avalanche luminescence is that PBR quantum states with broadband absorption are created in the material under intense laser irradiation. The PBR quantum states further increase the optical absorption in a wider spectral range, resulting in a positive feedback effect. The experimental results of time-resolved spectra and photoconductance reveal that the PBR is some different from the thermal radiation described by Planck’s law. Under strong light irradiation, the population of PBR levels in photoinduced blackbodies has two different population ways, the lighting population and the thermal population. When the thermal population process cannot completely cover up the lighting population process, the spectrum of the PBR will deviate from the description of Planck’s law. Due to the existence of lighting population, the color temperature of a photoinduced blackbody is higher than the temperature measured by a thermocouple. This discrepancy in temperature measurements also makes us speculate if the temperatures of stars (such as the Sun) obtained by spectral fitting need to be corrected.

Chemicals

All chemicals were of analytical grade and used without further purification. Yttrium oxide (Y₂O₃, 99.99%), Ytterbium oxide (Yb₂O₃, 99.99%) were supplied by Aladdin Reagent, Shanghai, China.

Preparation of thin sheet samples

Put 5 mmol powder sample into a tableting die with a diameter of 12 mm, pressurize it to 8 t using a BJ-15 tablet press (Tianjin Bojun Technology Co., Ltd., Tianjin, China), keep the pressure for 20 s, release the pressure and then take out the sheet sample (thickness about 2 mm).

Device characterization

The 808 nm (0 ~ 10 W) and 980 nm (0 ~ 30 W) power tunable CW laser diodes used in the experiments were produced by Beijing BWT Co., LTD. The 266 nm (0 ~ 120 mW), 405 nm (0 ~ 100 mW), 532 nm (0 ~ 500 mW), 650 nm (0 ~ 200 mW), and 1560 nm (0 ~ 15 W) power tunable CW lasers used in the experiments were produced by New Industry Optoelectronics Technology Co., Ltd., Changchun, China. The different parts of the steady-state luminescence spectra in different spectral regions were recorded using Ocean Insight HR4000CG-UV-NIR (200 nm ~ 1100 nm), YOKOGAWA AQ6370D (600 nm ~ 1700 nm), and Yokogawa AQ6375B (1200 nm ~ 2400 nm), respectively. The digital oscilloscope used in the experiment is DPO4104 (bandwidth 1 GHz, sampling rate 5 GS/s) Tektronix, USA. The spectrometer used in the measurements of time-resolved and transmission spectra is an 1 m monochromator SPEX 1000 M (HORIBA Jobin Yvon Inc., Edison, NJ, USA), equipped with a 600 lines/mm diffraction grating for NIR and visible fluorescence, a 1800 lines/mm diffraction grating for UV fluorescence, R928 photomultiplier tube for photoelectric conversion of UV fluorescence, H10300B-75 InGaAs photomultiplier tubes for photoelectric conversion of infrared fluorescence. The irradiance spectra of halogen lamps at 2856 K and 2432 K were calibrated by Beijing Institute of Metrology. The lock-in amplifier is Stanford SR830. The SourceMeter used for conductivity measurement is Keithley2450 (USA).

Measurement of transmission spectra of Yb₂O₃ thin slice irradiated by 980 nm laser

When the Yb₂O₃ powder enters the PBR state under the irradiation of 980 nm laser, it is not only a bright luminous body, but also an excellent optical absorber. In this case, the intense incandescence from the PBR greatly interfered with our measurement of transmission and absorption spectra. In order to avoid the interference, we built a detecting system for the measurement of transmission and absorption spectra based on a lock-in amplifier and an optical chopper, as shown in Supplementary Fig. 4. A halogen lamp with a broadband spectrum was utilized as the detection light source and a chopper was used to modulate the detection light. The modulated lamp light transmitted through the Yb₂O₃ thin sheet was detected and converted into a photocurrent signal by a photomultiplier tube (PMT) equipped on the 1000 M spectrometer. A lock-in amplifier was employed to amplify the transmission signal. Transmission spectra were recorded by using a PC.

Effect of heating on PBR threshold power

As shown in Supplementary Fig. 6, a layer of Y₂O₃ powder with a thickness of ~ 2 mm was coated on the surface of a Ni/Cr resistance wire, and the temperature of the sample was controlled by adjusting the heating current. When the Y₂O₃ powder was heated to 1500 K, under the irradiation of ~5 W 808 nm laser, the sample entered PBR state, indicating that electric heating has effectively reduced the threshold power of PBR. The scattering intensity of 808 nm laser was measured by a fiber optic spectrometer. Flame heating can also reduce the threshold power of PBR occurrence. Due to its similarity to electric heating experiments, relevant experiments are not described here.

Measurement of PBR spectra

To record the PBR spectra in the wide spectral range of 350 ~ 2500 nm, three optical fiber spectrometers in different wavelength ranges were used. The schematic diagram of the test system is shown in Supplementary Fig. 7. The irradiance spectra of the two halogen lamps were calibrated by Beijing Institute of Metrology and Inspection, as shown in Supplementary Fig. 13, and their spectra recorded by the test system is shown in Supplementary Fig. 14. The compensation factor for the three fiber spectrometers were obtained, respectively, by dividing their standard and measured spectra, as shown in Supplementary Fig. 15. We use the compensation factor measured by two bulbs with different color temperatures as a ratio, and the part whose ratio is closer to 1 is more reliable. Then, this test system was used to measure the PBR spectra of Y₂O₃ samples under different powers of 980 nm laser excitation. The PBR spectra before and after the correction of the compensation factor are shown in Supplementary Fig. 16. Finally, we spliced the spectra of different wavelength ranges together to obtain the PBR spectra of 350 ~ 2400 nm, as shown in Supplementary Fig. 18.

Measure the actual temperature of a PBR blackbody

Two different methods were used to measure the temperature of a sample in the PBR state. One is to get the color temperature of the sample by fitting the spectrum of PBR, and the other is to place a Pt/Rh thermocouple inside the powder sample to measure the actual temperature directly. To distinguish the two different measured temperatures, here we call the former the spectral temperature and the latter the actual temperature. To measure both temperatures simultaneously, a test system was established as shown in Supplementary Fig. 8. We placed a flat Pt/Rh thermocouple (diameter 2 mm, thickness 0.5 mm) in the middle of a cuboid (4×6×1.2) piled up by Yb₂O₃ powder, used a 980 nm laser to irradiate the sample above the thermocouple to excite its PBR (PBR spot diameter 5 mm). The PBR spectra were recorded with a fiber spectrometer in the spectral range of 350–1100 nm, and the actual temperature of the sample was read from the thermocouple thermometer. To ensure the accuracy of actual temperature measurement, we buried the thermocouple in the middle of the Yb₂O₃ powder sample, and focused the laser spot to cover the cuboid as much as possible, so that the whole thermocouple was in a stable and uniform temperature environment. To avoid direct heating of the thermocouple by the 980 nm laser, the thermocouple was completely embedded with the powder sample. However, even so, the PBR from the sample and the pump laser scattered by the sample inevitably heated the thermocouple, resulting in the practical effect of the laser heating the thermocouple. After the laser irradiation of a certain power for a period of time, the actual temperature measured by the thermocouple remained stable while the actual temperature value was read and the spectrum of the PBR was recorded. In addition, due to the possible distance between the PBR spot on the sample and the surface of the thermocouple, there may be a temperature gradient between them, which will result in a lower actual temperature. Even taking into account the above factors, there may be a deviation between the temperature measured by the thermocouple and the real temperature at the PBR spot. However, the temperatures we measured with the thermocouple are much lower than the spectral temperatures calculated from spectral fitting, as shown in Fig. 3b. It is unlikely that the temperature difference of >600 K between the spectral temperature and the actual temperature is due to the above deviation produced in the thermocouple measurements. On the other hand, the platinum-rhodium thermocouple showed no signs of melting in the temperature measurements, indicating that the real temperature of the sample did not reach the melting point (2045 K or 2239 K) of platinum or rhodium.

Measuring the temperature of the Pt/Rh thermocouple under the irradiation of 980 nm laser (0 ~ 30 W)

As shown in Supplementary Fig. 9, we focused 980 nm laser on one side of the flat Pt/Rh thermocouple to heat it up and induce intense thermal radiation. To reduce the influence of scattered laser on spectral measurement, we recorded the thermal radiation spectra of the Pt/Rh thermocouple from its backside with an optical fiber spectrometer. It can be seen that the spectral fitting temperature of the Pt/Rh thermocouple is almost same with the actual temperature measured by the thermocouple. As shown in supplementary Fig. 20, although the actual measured temperature is some higher than the spectral fitting temperature, the difference between them is <100 K for different irradiation powers.

Measurement of time-resolved spectra of the PBR

As shown in Supplementary Fig. 10, to measure the time-resolved spectra of Yb₂O₃ powder in PBR state, we used a 980 nm pulsed laser (peak power 30 W, repetition frequency 8 Hz, pulse width 20 ms) to pump the sample. The PBR from the sample was focused into the 1000 M spectrometer through a lens and the PBR signals were recorded by an oscilloscope. We obtained the time-resolved spectra of the PBR, as shown in Fig. 4b.

Measurement of photoconductivity of Yb₂O₃ powder in PBR state

As shown in Supplementary Fig. 11, to measure the photoconductivity of Yb₂O₃ powder in PBR state, two aluminum plates with a thickness of 2 mm were fixed on an alumina ceramic base. The gap between the two aluminum plates was 0.5 mm, and the positive and negative terminals of the source meter were connected to the two aluminum plates, respectively. We filled the gap with Yb₂O₃ powder and induced the PBR by a 980 nm laser from the top direction. The photoconductivity of Yb₂O₃ powder was measured under different laser power at a working voltage of 70 v.