2.1. Chemicals and reagents

Al₂O₃ with a particle size of 100–200 mesh was purchased from Sinopharm Chemical Reagent Co., Ltd. (China). Potassium iodate (KIO₃), potassium iodide (KI), sodium phosphate dodecahydrate (Na₃PO₄·12H₂O), and nitric acid (HNO₃) were of analytical grade; these were purchased from Sinopharm Chemical Reagent Co., Ltd. (China) and used as received without any further purification. Silver nitrate (AgNO₃) was chemically pure and was purchased from Beijing Chemical Factory. Radioiodine was in the form of Na¹³¹I aqueous solution and was obtained from HTA Co., Ltd. (China). Compressed Air and N₂ with 99.999% purity were purchased from Beijing Haikeyuanchang Practical Gas Co., Ltd. (China).

2.2. Instruments

The test system consisted of a KL-602 injection pump (Beijing KellyMed Co., Ltd., China), D08-1F gas flow controller (Beijing Sevenstar Electronics Co., Ltd., China), XMT618 temperature controller (Yuyao Changjiang Temperature Meter Instrument Factory, China), WRNK-191 temperature sensor (Shanghai Automation Instrumentation Co., Ltd., China), and custom-made tube furnace (Yangzhou Baoding Electric Equipment Factory, China).

The crystallinity of all the samples was evaluated by powder X-ray diffraction (XRD) measurements performed using a PANalytical X’pert Powder diffractometer, with Cu Kα radiation. The diffraction intensity data were collected in the 2θ range of 20° to 80° with a step length of 0.013° and a count time of 0.1 s per point at room temperature.

Thermogravimetric analysis-differential scanning calorimetry (TGA-DSC) was employed to evaluate the thermal stability of AgI and adsorbents. TGA-DSC was performed using a TA Q600 SDT system with 10 mg samples in alumina crucibles under N₂ or air. The temperature range employed was 25-800 ℃, with a heating rate of 10 ℃ min^-1 and a N₂ flow rate of 100 mL min^-1.

The specific surface areas of the samples were measured by nitrogen adsorption and desorption isotherms using a Micrometrics ASAP2010 system at 77 K. Prior to each analysis, the sample was degassed in vacuum at 300 °C for 2 h with pretreatment at 120 °C for 1 h.

X-ray photoelectron spectroscopy (XPS) analysis was performed using an Axis Ultra photoelectron spectroscope (Kratos Analytical Ltd.) with monochromatic Al Kα radiation (E = 1486.7 eV, source strength = 240 W) to confirm the composition of the adsorbent after calcination.

Inductively coupled plasma-atomic emission spectroscopy (ICP-AES, Leeman Science and Technology Ltd.) was used to confirm the actual Ag content in the adsorbents and the element constituting of the yellow powder after the iodine desorption test.

A microwave digestion system (MARS 6, CEM Corporation, United Kingdom) was used to digest the adsorbent.

2.3. Synthesis of Ag/Al₂O₃ adsorbent and Ag₃PO₄/Al₂O₃ adsorbent

Ag/Al₂O₃ adsorbent was prepared by a conventional impregnation method. The Ag content in the adsorbent was 10 wt.%, which is similar to that for the adsorbent used in our previous study ^{[14, 24]}. The as-synthesized adsorbent was designated as 10 wt.% Ag/Al₂O₃. AgNO₃ (1.6 g) was dissolved in 120 mL of deionized water, then 9.0 g of Al₂O₃ were added into the aqueous solution, and the contents were stirred at 80 ℃ in a lightproof environment for 24 h. Subsequently, water was removed from the solution by rotary evaporation at 70 ℃. The precursor of the adsorbent was then calcined in air at 450 ℃ for 2 h and ground in a mortar. The final adsorbent with 100-200 mesh was obtained after sieving.

10 wt.% Ag₃PO₄/Al₂O₃ adsorbent (the Ag content in this adsorbent is about 7.7 wt.%) was prepared by the following method: The precursor of the adsorbent, i.e., AgNO₃/Al₂O₃, was prepared by the same method as mentioned above. Then, 2 mL of sodium phosphate solution (Na₃PO₄) with a concentration of 50 g L^-1 was added into the flask to impregnate AgNO₃/Al₂O₃. Water was removed by rotary evaporation at 70 ℃. These two steps were repeated seven times until AgNO₃ was completely transformed to Ag₃PO₄. Finally, the adsorbent was washed several times with deionized water to remove NaNO₃ and excess Na₃PO₄, and then, it was dried and calcined in air at 450 ℃ for 2 h. The final adsorbent with 100-200 mesh was obtained after sieving.

The adsorbent (0.10 g) was digested with 10 mL of HNO₃ (65%) in the microwave digestion system. Digestion was carried out at 1000 W for 2 h. The solution was diluted with ultrapure water, and the concentration of Ag in the diluted solution was determined by ICP-AES ^{[14, 24]}. The results showed that the Ag/Al₂O₃ and Ag₃PO₄/Al₂O₃ adsorbents contained 9.8 ± 0.1 and 7.6 ± 0.1 wt.% Ag, respectively. The Ag contents were slightly smaller than the theoretical value.

2.4. Saturated adsorption and roast treatment of Ag₃PO₄/Al₂O₃ adsorbent

An ampoule with 1.0 g of 10 wt.% Ag₃PO₄/Al₂O₃ adsorbent was placed in a glass jar containing 2.0 g of solid iodine. This system was sealed and heated at 75 ℃ for 4 h ^[18]. The iodine concentration in this system was calculated by the saturated vapor pressure of iodine (about 1625 Pa at 348.15 K ^[6]) and was found to be about 0.56 mmol L^-1. The saturated adsorption sample was denoted as 10 wt.% Ag₃PO₄/Al₂O₃-SA.

The 10 wt.% Ag₃PO₄/Al₂O₃ adsorbent was roasted at 450, 550, 650, and 750 ℃ in air for 2 h to simulate the real process that the adsorbent would go through during the adsorption measurements. The sample was named 10 wt.% Ag₃PO₄/Al₂O₃-(T/t), where T indicates the calcination temperature (℃), and t is the calcination time (h).

The 10 wt.% Ag₃PO₄/Al₂O₃ adsorbent was also roasted at 750 ℃ for 2, 4, 6, and 8 h to optimize the longest time that it could remain stable at 750 ℃. The samples were also named following the same method mentioned above.

2.5. Synthesis of AgI/Ag and AgI/Ag₃PO₄ sample for thermogravimetric analysis

AgI and Ag₂CO₃ at a given mass ratio of 1:3.83 were mixed, ground, and roasted at 450 ℃ ^[26] for 2 h. The mixture was ground to obtain a homogeneous sample for TGA. The 25 wt.% AgI/Ag₃PO₄ sample was synthesized by a similar method.

2.6. Iodine adsorption and desorption

Adsorption of iodine on 10 wt.% Ag₃PO₄/Al₂O₃, 10 wt.% Ag/Al₂O₃, and Al₂O₃ adsorbents was tested at 450, 550, 650, and 750 ℃ under different carrier gases (N₂ and air) and gas flow rates (30 and 60 mL min^-1) using an apparatus designed by Cheng et al. ^{[14, 24]}. The apparatus consisted of three main parts, including a constant speed sampling system, a fixed bed adsorption system that could be used at high temperatures, and an effluent gas adsorption system, and 13X zeolite was employed to adsorb iodine at room temperature.

In Cheng’s experiment ^{[14, 24]}, radioactive iodine was injected into the system with petroleum ether. However, petroleum ether decomposes at high temperature, leading to the possible formation of organic iodine. In order to remove this interfering factor, an aqueous solution of iodine was selected as the iodine source. As the maximum solubility of iodine in water is about 30 mg/100 g, the aqueous solution of radioactive iodine contained radioiodine (~1 MBq) and stable iodine at a concentration of 180 μg mL^-1, and it was prepared using aqueous solutions of NaI¹³¹ and NaI¹²⁷. First, radioactive iodine solution (2 mL) was added into the sampling system. When injected into the fixed bed adsorption system with 1.20 g of the adsorbent loaded and then heated, the radioactive iodine solution turned into steam and passed through the adsorbent with N₂. Unadsorbed iodine was further adsorbed by 13X zeolite at room temperature in the effluent gas adsorption system. After the completion of the test, the radioactive count rates of the adsorbent in the fixed bed adsorption system and 13X zeolite in the effluent gas adsorption system were measured by a well counter to determine the decontamination factor (DF) and thus evaluate the performance of the adsorbents. The DF value is calculated by using equation (1) as follows:

where C_a is the radioactive count rate of adsorbent and C_e is the radioactive count rate of 13X zeolite. Every test was repeated three times, and the result was represented as average value ± standard deviation. The performance of the adsorbents can also be evaluated by Eq. (2) as follows:

After the iodine adsorption test at 200 ℃ under N₂ at a gas flow rate of 30 mL min^-1, the used adsorbent was tested at 750 ℃ under N₂ at a gas flow rate of 120 mL min^-1 to estimate the adsorption stability, by following the same method and using the same apparatus mentioned above. The amount of iodine desorbed from the adsorbent at 0.5, 1, 2, 4, 6, or 8 h was measured. Removal efficiency indicates the relative amount of radioiodine blown off the adsorbent and it can be used to estimate the stability of iodine.

3.1. Adsorption of radioactive iodine at high temperatures

Figure 1 shows the adsorption of ¹³¹I₂ by Al₂O₃, 10 wt.% Ag/Al₂O₃, and 10 wt.% Ag₃PO₄/Al₂O₃ adsorbents at different temperatures. The 10 wt.% Ag/Al₂O₃ and 10 wt.% Ag₃PO₄/Al₂O₃ adsorbents showed obvious advantages over Al₂O₃ with respect to the DF value and removal efficiency, as Al₂O₃ barely adsorbed the radioactive iodine. This result indicates that the adsorption ability of the adsorbent toward radioiodine is attributed to Ag, which is capable of reacting with iodine to form AgI.

Fig. 1Full-screen View

(Color online) Adsorption of ¹³¹I₂ by Al₂O₃, 10 wt.% Ag/Al₂O₃, and 10 wt.% Ag₃PO₄/Al₂O₃ adsorbents at different temperatures under carrier gas of N₂ (30 mL min^-1), with the y-axis showing log DF (a) and % removal efficiency (b).

Compared to the 10 wt.% Ag₃PO₄/Al₂O₃ adsorbent, the 10 wt.% Ag/Al₂O₃ adsorbent shows higher DF values at 450 and 550 ℃. Although the log DF values of the 10 wt.% Ag₃PO₄/Al₂O₃ adsorbent are more than 4 at 450 and 550 ℃, the 10 wt.% Ag/Al₂O₃ adsorbent can adsorb nearly 99.999% of the iodine input into the fixed bed adsorption system. The Ag content in the 10 wt.% Ag/Al₂O₃ adsorbent is higher than that in the 10 wt.% Ag₃PO₄/Al₂O₃ adsorbent, where Ag exists in the form of silver ions at about 7.7 wt.% concentration. It is noteworthy that the Ag content can affect the DF value; therefore, the results are reasonable. However, when the temperature increases from 550 to 650 ℃, the DF value of the 10 wt.% Ag/Al₂O₃ adsorbent decreases sharply by three orders of magnitude. When the temperature increases further to 750 ℃, the 10 wt.% Ag/Al₂O₃ adsorbent suffers substantial loss in adsorption ability. In contrast, the DF value of the 10 wt.% Ag₃PO₄/Al₂O₃ adsorbent is very stable at different temperatures. The 10 wt.% Ag₃PO₄/Al₂O₃ adsorbent shows the same performance at 450, 550, and 650 ℃, and the DF value only decreases slightly; this material can still adsorb more than 99% of the radioactive iodine at 750 ℃. Adsorbents that can be used at such a high temperature have never been reported in the literature. However, undeniably, much more systematic exploration and evidence are demanded to identify the reasons for the higher removal efficiency of the 10 wt.% Ag₃PO₄/Al₂O₃ adsorbent compared to that of the 10 wt.% Ag/Al₂O₃ adsorbent.

The results of the adsorption tests on 10 wt.% Ag₃PO₄/Al₂O₃ under different conditions are shown in Fig. 2.

Fig. 2Full-screen View

(Color online) Adsorption of ¹³¹I₂ by 10 wt.% Ag₃PO₄/Al₂O₃ adsorbent under different carrier gases and gas flow rates.

Fig. 2 demonstrates that when the carrier gas is changed from N₂ to air or when the gas flow rate is increased from 30 to 60 mL min^-1, the DF value of the 10 wt.% Ag₃PO₄/Al₂O₃ adsorbent hardly declines at 450, 550, and 650 ℃. Although the DF value declines slightly with an increase in the gas flow rate from 30 to 60 mL min^-1 at 750 ℃, the material can still adsorb more than 99% of the radioactive iodine at 750 ℃.

3.2. Characterization of the adsorbents

Fig. 3 shows the nitrogen adsorption–desorption isotherms of the adsorbents, indicating that 10 wt.% Ag₃PO₄/Al₂O₃ is a mesoporous material, with a pore diameter distribution similar to that calculated by using the Barrett–Joyner–Halenda method. The specific surface area of the untreated adsorbent was 135 m² g^-1, and it was largely stable through high-temperature calcination. Calcination at high temperatures did not affect the structure of the adsorbent. However, it resulted in a slight decrease in the Brunauer–Emmett–Teller (BET) specific surface area and a minor increase in the average pore diameter, as indicated by the values listed in Table 1.

Fig. 3Full-screen View

(Color online) Nitrogen adsorption–desorption isotherms (a) and pore size distribution (b) of 10 wt.% Ag₃PO₄/Al₂O₃ adsorbents roasted at different temperatures.

With an increase in the calcination temperature, the BET specific surface area decreases and the adsorption average pore diameter increases. After roasting at 750 ℃ for 2 h, the specific surface area of the adsorbent exhibits about 30% decrease. The falloff in the BET specific surface area can partly result in a decline of the DF value, since a lower specific surface area generally indicates a lower reaction area. On the contrary, the average pore diameter of the adsorbent increases from 6.34 to 8.42 nm after roasting at 750 ℃ for 2 h. The larger the pore diameter, the greater is the pore diffusivity. Hence, the structural change has two conflicting effects on the DF value. Considering that when the temperature increases from 650 to 750 ℃, the specific surface area decreases by 10% and the DF value decreases far more than 10%, the structural change is unlikely to be the main reason for the decrease in the DF value.

After the saturated adsorption experiment, the color of the adsorbent changed from pale yellow to brown. Fig. 4a shows that the diffraction peaks of Ag₃PO₄ (JCPDS no. 06-0505) in the adsorbent are replaced by those of AgI (JCPDS no. 09-0374). This confirms that iodine can react with Ag₃PO₄ to form AgI. Thus, the adsorption mechanism for the 10 wt.% Ag₃PO₄/Al₂O₃ adsorbent is chemical adsorption instead of physical adsorption, since iodine can react with silver phosphate. Although the specific reaction mechanism is unconfirmed, the possible reactions are shown in Eqs. (3) and (4) ^[22] as follows:

Fig. 4Full-screen View

(Color online) Powder XRD patterns of (a) 10 wt.% Ag₃PO₄/Al₂O₃ and 10 wt.% Ag₃PO₄/Al₂O₃-SA; (b) 10 wt.% Ag₃PO₄/Al₂O₃ adsorbents roasted at different temperatures; and (c) 10 wt.% Ag₃PO₄/Al₂O₃ adsorbents roasted at 750 ℃ for different lengths of time.

Fig. 4b shows the XRD patterns with sharp diffraction peaks corresponding to Ag₃PO₄ and minor diffraction peaks corresponding to Al₂O₃ (JCPDS no. 04-0880). With an increase in temperature, the intensity of the diffraction peaks of Ag₃PO₄ reduces gradually. This can be attributed to the decrease in the Ag₃PO₄ content or the better dispersion of Ag₃PO₄. As the width of the diffraction peaks of Ag₃PO₄ does not increase after calcination, the possibility of its decomposition increases. Furthermore, the diffraction peaks of Ag, the decomposition product of Ag₃PO₄, cannot be observed in the XRD patterns due to its small amount. The increase in the sharpness of the diffraction peaks of Al₂O₃ at higher calcination temperatures indicates an improvement in crystallinity. The increase in crystallinity leads to a decrease in the specific surface area of Al₂O₃ (Table 1). This result can thus verify that the decrease in both the specific surface area and the Ag₃PO₄ content may lead to a decrease in the DF value of the adsorbent. However, when the temperature increases from 450 to 650 ℃, the diffraction peak intensity of Ag₃PO₄ changes drastically, and the DF value of the adsorbent decreases slightly. The decomposition of Ag₃PO₄ has little effect on the DF value under the test conditions. Moreover, the decline of the diffraction peak intensity of Ag₃PO₄ is less than the decrease in the DF value when the temperature increases from 650 to 750 °C. Therefore, some other reasons might be mainly responsible for the decrease in the DF value at 750 ℃.

Fig. 4c shows XRD patterns indicating that with an increase in the calcination time, the diffraction peaks of Ag₃PO₄ become weaker. When calcination at 750 ℃ was prolonged for 8 h, the diffraction peaks of Ag₃PO₄ became very weak, indicating a low Ag₃PO₄ content.

XPS measurements were carried out to confirm the composition of the adsorbent after calcination. Fig. 5a shows obvious O 1s, Ag 3d, P 2p, and Al 2p peaks, thus confirming the existence of Ag, O, P, and Al elements. The Ag 3d peaks shown in Fig. 5b were analyzed to confirm the chemical status of Ag. The peaks of Ag 3d_5/2 and Ag 3d_3/2 were located at ~368 and ~374 eV. After calcination at high temperature, in particular, at 650 and 750 ℃, the peaks corresponding to Ag 3d_5/2 and Ag 3d_3/2 slightly shifted to higher binding energy, indicating the possible change in the chemical status of Ag. Fig. 5c demonstrates that the Ag 3d_5/2 and Ag 3d_3/2 peaks of the 10 wt.% Ag₃PO₄/Al₂O₃-(750/2) adsorbent can be divided into two different peaks at 368.0, 368.5 eV; and 374.0, 374.5 eV. According to related reports ^[27], the peaks at 368.5 and 374.5 eV should be attributed to Ag⁰, and the peaks at 368.0 and 374.0 eV should be attributed to Ag⁺ in Ag₃PO₄. This can prove the existence of metallic Ag and the decomposition of Ag₃PO₄ caused by high-temperature calcination. Moreover, this result is in agreement with the inference from Fig. 4.

Fig. 5Full-screen View

(Color online) X-ray photoelectron spectra of (a) 10 wt.% Ag₃PO₄/Al₂O₃ adsorbent; (b) 10 wt.% Ag₃PO₄/Al₂O₃ adsorbents roasted at different temperatures; and (c) 10 wt.% Ag₃PO₄/Al₂O₃-(750/2).

3.3. Stability

Both the reaction dynamics factor ^[28] and the thermodynamic stability of AgI can affect the DF value. TGA-DSC analysis of 25 wt.% AgI/Ag and 25 wt.% AgI/Ag₃PO₄ was carried out to evaluate the thermal stability of AgI in different adsorbents. The melting point of the solid solution with a large amount of AgI is low. Moreover, if the mass fraction of AgI is too small, TGA-DSC analysis may not be sensitive enough to detect the small amount of AgI desorbed from the sample. Therefore, the mass fraction of AgI was selected to be 25 wt.%. The results of the TGA-DSC analysis are shown in Fig. 6.

Fig. 6Full-screen View

(Color online) TGA-DSC curves of 10 wt.% Ag₃PO₄/Al₂O₃ adsorbent (a, under N₂; b, under air), 25 wt.% AgI/Ag (c, under N₂), and 25 wt.% AgI/Ag₃PO₄ (d, under N₂).

The adsorbent exhibits continuous loss in weight during the measurements. In the temperature range 25 to 400 ℃, the mass loss was about 4.6%, and this should be attributed to the desorption of water ^[29]. Furthermore, the DSC curve shows the existence of a small endothermic peak at around 521 ℃, which may be caused by the decomposition of Ag₃PO₄. Therefore, the mass loss in the range 521 to 750 ℃, which is about 0.6%, should be attributed to the decomposition of Ag₃PO₄. In the same temperature range, the mass loss of the adsorbent under air atmosphere is about 0.7%, as shown in Fig. 6b. Oxidation or greater degradation of Ag₃PO₄ in the presence of air can explain the difference. This might be why the adsorbent shows better performance under N₂ atmosphere in the iodine adsorption test. Integration of the results shown in Figs. 4 and 5 indicates that long-time calcination at a high temperature can lead to deactivation of the 10 wt.% Ag₃PO₄/Al₂O₃ adsorbent.

25 wt.% AgI/Ag is stable, with nearly no mass loss, after heating to 600 ℃. With the continuous increase in temperature, mass loss occurs at an accelerating rate, which corresponds to the desorption and decomposition of AgI ^{[30, 31]}. The total mass loss in the temperature range 25 to 750 ℃ is 1.5%. This explains the main reason for the low DF values of 10 wt.% Ag/Al₂O₃ at 650 ℃, and in particular, at 750 ℃. The DSC curve shows an endothermic peak at 554 ℃ due to the melting of AgI.

25 wt.% AgI/Ag₃PO₄ is extremely stable in the temperature range 25 to 750 ℃, with only a small mass increase which may be attributed to some side reactions ^[32]. AgI is stable in AgI/Ag₃PO₄ because it can form a solid solution with Ag₃PO₄ ^[25]. The formation of a solid solution can increase the melting point of AgI depending on the mass ratio of AgI and Ag₃PO₄. The DSC curve shows two endothermic peaks at 325 and 491 ℃, which are due to the phase transformations of the AgI/Ag₃PO₄ solid solution.

To confirm the formation of a solid solution between AgI and Ag₃PO₄, XRD data of 25 wt.% AgI/Ag₃PO₄-unroasted and 25 wt.% AgI/Ag₃PO₄-roasted were collected.

After calcination at 200 °C for 2 h, the diffraction peaks of AgI disappeared, and several new diffraction peaks ^{[33, 34]}, which are attributed to (AgI)_x(Ag₃PO₄)_y (shown in Fig. 7a) could be observed. In addition, the diffraction peaks of Ag₃PO₄ shifted to the high-angle region. During the forming of the solid solution between AgI and Ag₃PO₄, diffusion of the smaller I^- ion into the Ag₃PO₄ lattice led to a smaller crystal space, causing the diffraction peaks to shift to the high-angle region ^[35], as shown in Fig. 7b.

Fig. 7Full-screen View

(Color online) Powder XRD patterns of 25 wt.% AgI/Ag₃PO₄-unroasted and 25 wt.% AgI/Ag₃PO₄-roasted.

When the temperature exceeds the melting point of AgI, AgI becomes unstable. AgI could decompose and be blown off the absorbent ^{[30, 31]}. Therefore, a traditional Ag adsorbent cannot be used above the melting point of AgI. Thus, the 10 wt.% Ag/Al₂O₃ adsorbent synthesized in this study showed a performance similar to that of Cheng’s Ag impregnated Al₂O₃ adsorbent ^[14]. When the test temperature was increased to 750 ℃, both these Ag-loaded adsorbents lost their adsorption ability. In contrast, the stability of AgI on the 10 wt.% Ag₃PO₄/Al₂O₃ adsorbent was promoted because of the formation of a solid solution between AgI and Ag₃PO₄, and the Ag₃PO₄/Al₂O₃ adsorbent exhibited significantly higher DF values than did Ag/Al₂O₃ at 650 and 750 ℃. However, 750 ℃ is close to the limiting temperature at which that the solid solution can exist ^[25]. AgI becomes unstable at this temperature. As a result, the DF value of the 10 wt.% Ag₃PO₄/Al₂O₃ adsorbent at 750 ℃ would not be as large as that at temperatures below 650 ℃. Nonetheless, it satisfies the Environmental Radiation Protection Standards for Nuclear Power Operations ^[36].

In order to examine the thermal stability of the adsorbent under the test conditions, an iodine desorption test was conducted on the 10 wt.% Ag₃PO₄/Al₂O₃ adsorbent; the corresponding results are shown in Fig. 8. From this figure, iodine reserved was calculated using equation (2) mentioned above. With an increase in desorption time from 0.5 to 4 h, the amount of iodine desorbed from the adsorbent increased slightly. However, it was still less than 0.5% of the total iodine adsorbed by the adsorbent. When the desorption time increased to 6 h, the amount of iodine desorbed was nearly 1%. More than 6% of the total iodine desorbed from the adsorbent after it was roasted at 750 ℃ for 8 h. Fig. 4c shows the transformation of most of the Ag₃PO₄ into Ag after calcination for 8 h at 750 ℃. Thus, when Ag₃PO₄ was totally transformed into Ag, the adsorbent would deliver similar performance as the 10 wt.% Ag/Al₂O₃ adsorbent.

Fig. 8Full-screen View

Thermal stability of ¹³¹I₂ in 10 wt.% Ag₃PO₄/Al₂O₃ adsorbent at 750 ℃.

After the iodine desorption test on 10 wt.% Ag/Al₂O₃-SA, a yellow powder was observed in the effluent gas adsorption system. This powder was dissolved in KI solution ^[37] and measured by ICP-AES. The results indicated that it contained Ag; thus, the yellow powder could be AgI. Moreover, at the same time, greater desorption of molecular iodine was observed. The results clearly indicated blowing off of AgI from the adsorbent, followed by decomposition under the test conditions. This result is in agreement with the conjecture obtained from Fig. 6. Notably, the DF value of the adsorbent at 750 ℃ mainly depended on the stability of AgI.