Aluminum silicate deposition on polyethylene naphthalate films and its application to Cs absorber in Cs aqueous solution Open Access

Aluminum silicate is a main component of zeolite minerals. It is explained that negatively charged Al atoms surrounded by SiO₂ attract cations onto the surface in aqueous solutions.^1–4 Therefore, zeolite minerals are used as cesium absorbers in aqueous solutions.^5–7 Since the 2011 accident at the Fukushima Daiichi Nuclear Power Plant, there has been growing interest in the study of nuclear accidents and their environmental and health impacts. The zeolite minerals are used as absorbers for radioactive Cs in the wastewater although they become heavy radioactive waste of used zeolite. If aluminum silicate flexible films could be realized as cation absorbers for Cs sorption, a Cs absorber with a roll-to-roll cycling system could be developed, which might reduce the nuclear waste generated from used zeolite. In our group, therefore, we have examined the aluminum silicate deposition on flexible films.

In our previous work, Mori et al.⁸ demonstrated room-temperature atomic layer deposition (RTALD) of aluminum silicate on Si with an Al precursor of trimethyl aluminum (TMA) and a Si precursor of tris(dimethylamino)silane (TDMAS). The oxidizing gas was plasma-excited humidified argon. It was also effective in leaving the hydroxyl passivation⁹ after oxidization, which promotes the chemical reaction related to the deposition at room temperature (RT). To achieve the mixture of aluminum and silicon oxides with the atomic level, a mixed saturation layer of TMA and TDMAS was formed in the adsorption step by the sequential adsorption where the TDMAS was introduced to the growing surface first, followed by the TMA adsorption at RT.⁸ Since the TMA molecule adsorbs on the surface to extract the adsorbed TDMAS molecule,⁸ the atomic content of Al to Si in the deposited film was controlled by the TMA exposure. The aluminum silicate film exhibited an Na sorption capacity in the NaCl solution and confirmed its exchange to potassium in the KCl aqueous solution, suggesting the applicability of the film as a cation absorber in aqueous solutions.

Building on previous work, we extend our examination of aluminum silicate deposition on a flexible film of polyethylene naphthalate (PEN). This examination is conducted to confirm the applicability of the present film as a roll-to-roll cyclic cesium (Cs) absorber. The deposited film exhibits Cs sorption capacity in a CsCl aqueous solution and its exchange with Na by immersing the film in the NaCl solution. Reverse exchange is also confirmed by immersing the film in the CsCl solution again. In this paper, the related experiments and Cs elimination test are presented using a prototype roll-to-roll system with the aluminum silicate-deposited film.

The aluminum silicate films were deposited on PEN films (Teijin Q51-A4) with a thickness of 100 μm. The film size was 297 × 210 mm². Before deposition, the surface of the PEN film was treated with the plasma-excited humidified argon in the ALD chamber for a prolonged period for cleaning and initial surface oxidization. The RTALD system is shown in Fig. 1. A stainless-steel reaction chamber was connected to a turbo molecular pump (UTM-150, ULVAC) and a gas introduction system. TMA and TDMAS were used as precursors of Al and Si, respectively. The plasma-excited humidified argon as the oxidizing gas was generated with an Ar carrier gas using a water bubbler and a solenoid coil. The bubbler temperature was 50 °C. The plasma excitation was made with a frequency of 13.56 MHz and an RF power of 100 W. The ALD process was conducted at RT by repeating adsorption, evacuation, oxidization, and evacuation. In the adsorption step, the TDMAS gas was introduced with an exposure of 2.0 × 10⁵ l, where 1 l corresponds to 1.33 × 10⁻⁴ Pa s. Then, TMA was introduced with a limited exposure of 150 l. In the first adsorption, the surface was saturated with TDMAS molecules, followed by TMA attaching to the surface by extracting a certain amount of the surface TDMAS molecules. Then, the chamber was evacuated for 70 s. The plasma-excited humidified argon was introduced to the ALD chamber with a flow rate of 5 SCCM for 300 s. Then, the chamber was evacuated for 30 s. In this experiment, the ALD cycle number was 100.

The chemical composition of the deposited film and its adsorbates was evaluated by x-ray photoelectron spectroscopy (ULVAC PHI model 1600) with an MgKα x-ray source with an energy of 1253.6 eV. The take-off angle of photoelectrons with respect to the sample surface was 90°. The film morphology was confirmed by an atomic force microscope (Nanosurf NaioAFM). The film thickness was measured using a spectroscopic ellipsometer (JASCO M-220).

In the present study, the ion sorption capacity of the aluminum silicate-deposited PEN film was examined by measuring Cs and Na densities on the surface using XPS where the sample surface was immersed in CsCl and NaCl aqueous solutions. The concentration of CsCl and NaCl was 10 mM. Before the XPS measurement, the film surface was dried by N₂ blow. The remaining Cs in the CsCl solution was also measured by XPS where a 0.1 ml solution was dropped uniformly with a surface tension on a 8 × 8 mm² SiO₂ sample and the surface was dried in air.

To confirm the cyclic adsorption of Cs in CsCl aqueous solution, a prototype roll-to-roll system was developed, as shown in Fig. 2. In this system, an aluminum silicate-deposited PEN tape with a width of 2 cm and a loop length of 125 cm was used. To fabricate the loop tape, we prepared a PEN film with a size of 210 × 297 mm² and placed it in the reaction chamber. At that time, the PEN film was curled and positioned on the bottom plate of the sample holder. Then, ALD was performed as described above. To obtain the loop tape, the PEN film was cut into tapes with a width of 2 cm, and the tapes were connected to each other using small pieces of the Kapton adhesive tape. The loop tape was dipped in the CsCl solution followed by being dipped in the NaCl solution, where we assumed the Cs ions were adsorbed to the tape surface and dropped in the concentrated NaCl solution. The loop tape speed was 6.5 cm/s. The CsCl concentration was 1.66 × 10⁻⁵ M. The solution volume and temperature were 60 ml and 18 °C, respectively. The NaCl concentration was 1 M with the solution volume and temperature being 60 ml and 18 °C, respectively. To ensure the perfect exchange from Cs to Na, the higher concentration of 1 M was chosen. The length of the dipped part was 5 cm in each of the CsCl and NaCl solutions.

The deposition of aluminum silicate on PEN films was examined. Since the growth per cycle measured from the Si surface was 0.098 nm/cycle, the total thickness of the film was calculated as 9.8 nm. The Si to Al atomic ratio was approximately 3. Images of the control sample of the untreated PEN film, aluminum silicate-deposited PEN film, and its AFM image are shown in Figs. 3(a)–3(c), respectively. The thickness of the aluminum silicate was estimated at 9.8 nm. The sample size was around 2 × 2 cm². From the appearance of the films, the difference between untreated- and aluminum-silicate-deposited PEN films was unobservable. From the AFM image, the average surface roughness of aluminum silicate-deposited PEN film was measured to be 0.76 nm, while the initial surface roughness of the PEN film was 1.04 nm. This suggests that the present RTALD allows for uniform deposition while maintaining the initial surface roughness. An XPS survey spectrum measured from the aluminum silicate-deposited film is shown in Fig. 4. The surface was etched by an Ar bombardment for 30 min to remove the adsorbate from the air. The peaks related to Al, Si, and O were confirmed, suggesting the deposition of an aluminum silicate film on the PEN film. The Al to Si atomic ratio was estimated at ∼3.0 from the peak ratio of Al 2p and Si 2p.

To confirm Cs and Na sorption capacities in CsCl and NaCl solutions, the aluminum silicate-deposited PEN films were immersed in an NaCl aqueous solution, followed by being dipped in a CsCl aqueous solution. The XPS spectra are shown in Fig. 5. In this experiment, the ALD cycle number was 100, which corresponds to a deposition thickness of 9.8 nm. The NaCl concentration, volume, and temperature were 10 mM, 20 ml and 19 °C, respectively. The CsCl concentration, volume, and temperature were also 10 mM, 20 ml, and 19 °C, respectively. The dipping time was 30 min for each of the CsCl and NaCl solutions. As shown in Fig. 5, when the film was immersed in the NaCl solution, the Na 1s peak emerged, indicating Na adsorption on the surface. The Na density was calculated as $3.73× 10 14 atoms / c m 2$⁠. Then, when the film was immersed in the CsCl solution, the Na 1s peak diminished and the Cs 3d peak emerged. This indicates that the surface Na was replaced with Cs. The surface Cs density was calculated to be 2.08 × 10¹⁵ atoms/cm². We also checked the reverse sorption as shown in Fig. 6. The experiments mentioned above indicate that sorption capacities for Na and Cs cations in the NaCl and CsCl aqueous solutions and the adsorbates are exchanged through solution exchange. It is considered that the cyclic sorption of Cs from CsCl aqueous solution can be achieved by repeating the CsCl and NaCl solution dipping.

To confirm the elimination of Cs from the CsCl aqueous solution, we repeated the dipping the aluminum silicate-deposited PEN film in CsCl and NaCl solutions for 30 s alternately and checked the concentration change of Cs in the remaining CsCl aqueous solution. In this experiment, we used a film with a size of 20 × 20 mm² with both sides of the film covered with aluminum silicate. Here, the CsCl concentration, the solution temperature, and its volume were 1.66 × 10⁻³ M, 18 °C, and 20 ml, respectively. The NaCl concentration, the solution temperature, and its volume were 1 M, 18 °C, and 20 ml, respectively. Before the XPS measurement, the surface was dried in the air. As shown in Fig. 7, the Cs peak was confirmed before dipping although it decreased after the 30 cycles of repetition. If we assume that the Cs peak height is linear with the Cs concentration, this test implies the removal of Cs from the Cs aqueous solution. The conveyance of removed Cs to the NaCl solution was also confirmed by XPS as shown in Fig. 8. In this experiment, after the 30 cycles of dipping, the Cs 3d 5/2 peak around 725.8 eV was confirmed from the remaining NaCl solution.

To confirm the applicability of the present technique to a roll-to-roll cycle system, the Cs elimination test was performed with the prototype system shown in Fig. 2. The Cs concentration in the remaining solution was confirmed by XPS. As shown in Fig. 9, the Cs 3d peak diminished with a treatment time of 4 min or more. We consider that the experimental results strongly suggest an applicability of the roll-to-roll type Cs absorber. For practical application, more elaborate investigations about the sorption capacities as a function of Cs and Na concentrations must be necessary to design the system based on the binding constants of Cs and Na on the RTALD aluminum silicate.

In the present paper, we demonstrated the aluminum silicate deposition on the flexible films of PEN. Looking back elsewhere, Yan and Bein^10,11 reported the complex film consisting of CaA zeolite crystals and binders. They prepared the film type zeolite on a quartz crystal microbalance and confirmed the sorption capacities to cations of Na⁺, K⁺, and Rb⁺, suggesting a possibility as a film type absorber. However, they calcined the film at 400 °C for the stabilization, which might be a problem for the fabrication on the flexible film. The present low-temperature fabrication process is considered to have a merit for realizing a flexible cycling system. They demonstrated a tunable molecular sieve function with pores in the zeolite crystal. In the present film, the film is not crystallized due to the RT process. To realize the molecule sorption selectivity in the present absorber, the low-temperature crystallization using a plasma shower is examined, which will be released elsewhere.

Aluminum silicate deposition on a flexible PEN film was demonstrated using RTALD with the sequential adsorption. TDMAS and TMA were sequentially introduced to the sample surfaces to achieve the mixed saturation layer. The oxidizing gas was plasma-excited humidified argon. The aluminum-silicate-deposited film exhibited sorption capacities for Na and Cs cations in NaCl and CsCl aqueous solutions and the adsorbates were exchanged through solution exchange. To realize a Cs ion elimination system from Cs-containing water, a roll-to-roll system was proposed using aluminum silicate-deposited PEN films. The Cs concentration in a CsCl aqueous solution was confirmed to be reduced by this treatment. The experiments suggested the applicability of the film as a treatment system for radioactive-contaminated water in accident-affected atomic power plants.

This work was partly supported by JSPS KAKENHI under Grant Nos. 23H00098 and 22K18787.

The authors have no conflicts to disclose.

Hibiki Takeda: Data curation (lead); Formal analysis (lead); Methodology (equal); Writing – review & editing (equal). Haruto Suzuki: Data curation (equal); Methodology (equal); Writing – review & editing (equal). Ryo Miyazawa: Data curation (equal); Writing – review & editing (equal). Masanori Miura: Methodology (equal); Writing – review & editing (equal). Bashir Ahmmad: Writing – review & editing (equal). Fumihiko Hirose: Conceptualization (lead); Formal analysis (lead); Funding acquisition (lead); Supervision (lead); Writing – original draft (lead).

The data that support the findings of this study are available from the corresponding author upon reasonable request.