2022 Volume 62 Issue 5 Pages 871-874
In this study, a method was presented for detecting low concentrations of elements in a high purity water sample using a portable total reflection X-ray fluorescence (TXRF) spectrometer. Preparing the dry residue of a sample droplet with large volume on a hydrophobic film coated sample holder was effective for improving the detection limit expressed as the concentration of a target element in a sample solution. When a TXRF spectrum of the dry residue of a 200 µL droplet of a solution containing 10 µg/L of Cr that was prepared by diluting a commercially available 1000 mg/L Cr standard solution with ultrapure water on a hydrophobic film coated sample holder was measured, a detection limit of 0.13 µg/L was achieved for Cr.
Total reflection X-ray fluorescence (TXRF) analysis has been used for trace elemental analysis as described elsewhere,1) and detection limits for many elements obtained by a TXRF spectrometer with an X-ray tube are typically in the pg range. Detection limits in the fg range were achieved by using synchrotron radiation.2,3,4) In order to perform trace elemental analysis of a solution sample using TXRF analysis, a droplet of the solution sample is needed to be dried on a sample holder whose surface is flat and smooth enough to reflect the incident X-ray beam with high X-ray reflectivity. A quartz glass substrate is usually used as a sample holder, and a few tens of μL of a sample solution is usually placed on a sample holder. In the case that a detection limit for an element obtained from a spectrum of the dry residue of a 10 μL droplet of a sample solution is 1 pg, a detection limit expressed as the concentration of the element in the sample solution is 100 ng/L. In TXRF analysis, by increasing the volume of a droplet of a sample solution placed on a sample holder tenfold, the net intensity of a fluorescent X-ray peak originating from an element in the dry residue is expected to increase tenfold. This is because the mass of the element in the dry residue increases tenfold. Therefore, increasing the volume of a sample droplet makes it possible to detect lower concentrations of elements in a sample solution. However, the size of the dry residue tends to be large when a sample droplet with large volume is dried. The ratio of fluorescent X-ray photons reaching the X-ray detector to those emitted from the dry residue will decrease with an increase in the size of the dry residue. Therefore, increasing the volume of a droplet of a sample solution placed on a sample holder without any plans may not lead to the expected increase in the net intensity of a fluorescent X-ray peak. In order to significantly improve the detection limit expressed as the concentration, the size of the dry residue of a large volume droplet of a sample solution is needed to be reduced. When a droplet of a sample solution is placed on a sample holder having a hydrophobic surface, the spread of the droplet can be reduced. Therefore, the use of such a sample holder can lead to a reduction in the size of the dry residue of a sample solution. For example, the use of a diamond-like carbon (DLC) film coated quartz glass substrate reduced the size of the dry residue of a sample solution compared with the use of a quartz glass substrate because the surface of the DLC film was more hydrophobic than that of the quartz glass substrate.5) Therefore, the detection limit expressed as the concentration can be significantly improved by preparing the dry residue of a large volume droplet of a sample solution on a sample holder having a hydrophobic surface.
A portable TXRF spectrometer that was developed in 20066) has been improved,7) and a Cr detection limit of 8 pg was obtained when a TXRF spectrum of the dry residue of a 1 μL droplet of a solution containing 1 mg/L of Cr was measured in a low vacuum.8) This portable spectrometer can be applied to on-site analysis of metals eluted from stainless steel products into high purity water for evaluating the corrosion resistance of them in the production process. For example, high purity water used as cooling water passes through stainless steel pipes in a nuclear power plant with a boiling water reactor,9) and these pipes are needed to be protected from corrosion deterioration caused by water. Kawamura et al.10) reported that when an eluate prepared by soaking a stainless steel fork in 36 mL of ultrapure water at 60°C for 30 min was analyzed by using graphite furnace atomic absorption spectrometry, the mean value of the concentration of Cr was 6 ppb (μg/L). The detection limit should be in the ng/L range in order to reduce the variation in the quantitative value of a metallic element in an eluate from a stainless steel product whose concentration is several μg/L. Therefore, a method for detecting low concentrations of elements in a high purity water sample using the portable TXRF spectrometer is needed to be developed in order to apply this spectrometer to the evaluation of the corrosion resistance of stainless steel products. In this study, we investigated the effectiveness of measuring the dry residue of a large volume droplet of a very dilute aqueous solution sample on a hydrophobic film coated sample holder for improving the detection limit of the portable TXRF spectrometer expressed as the concentration. This study showed that a detection limit of one hundred and several tens of ng/L was achieved for Cr when a spectrum of the dry residue of a 200 μL of an aqueous solution containing 10 μg/L of Cr on a hydrophobic film coated sample holder was measured in air for 600 s.
A solution containing 10 mg/L of Zn and that containing 5 mg/L of Zn were prepared from a 100 mg/L Zn standard solution (Zn(NO3)2 in 0.1 mol/L HNO3) (FUJIFILM Wako Pure Chemical Co., Osaka, Japan) and distilled water. A solution containing 10 μg/L of Cr was prepared from a 1000 mg/L Cr standard solution (K2Cr2O7 in 0.1 mol/L HNO3) (FUJIFILM Wako Pure Chemical Co., Osaka, Japan) and ultrapure water (Kanto Chemical Co., Inc., Tokyo, Japan). In this study, the following sample holders were used: quartz glass substrate (hereinafter referred to as quartz glass sample holder); DLC film coated quartz glass substrate (hereinafter referred to as DLC sample holder); hydrophobic film coated DLC sample holder. The size of each quartz glass sample holder was 30 mm in diameter and 5 mm in thickness. The size of each quartz glass substrate used for preparing DLC sample holders was 30 mm in length, 30 mm in width, and 5 mm in thickness. DLC film coating was conducted by Nanotec Co. (Kashiwa, Japan), and thicknesses of DLC films were set to 1 μm. A commercially available waterproof spray for shoes containing fluorine resin as a hydrophobic agent was used for preparing a hydrophobic film coated DLC sample holder. By spraying the product into a plastic centrifuging tube, liquid containing the hydrophobic agent was obtained. Then a 600 μL droplet of the liquid containing the hydrophobic agent was spin coated on a DLC sample holder. Hydrophobic film coating was conducted for improving the water repellency of the DLC sample holder. All quartz glass substrates used in this study were purchased from Sigmakoki Co., Ltd. (Hidaka, Japan), and the flatness of each quartz substrate was λ/20 (λ = 632.8 nm). One μL droplets of the solution containing 10 mg/L of Zn and the solution containing 5 mg/L of Zn were placed and dried on quartz glass sample holders. The drying of the droplets of these sample solutions was conducted by using an electric hot plate. 200 μL droplets of the solution containing 10 μg/L of Cr were placed and dried on a DLC sample holder and a hydrophobic film coated DLC sample holder. A 200 μL droplet of ultrapure water was placed and dried on a hydrophobic film coated DLC sample holder. The droplets of the sample solution containing 10 μg/L of Cr and ultrapure water were dried by heating them using a Peltier device.
The setup of the portable TXRF spectrometer used in this study was almost similar to that reported in a previous paper,11) but a silicon drift detector (SDD) VITUS H30 (Ketek GmbH, Munich, Germany) having an active area of 30 mm2 was used in this study instead of an SDD VITUS H7 (Ketek GmbH) with an active area of 7 mm2 used in the previous paper. The SDD VITUS H30 was used for all TXRF measurements except the measurement of the solution containing 10 mg/L of Zn. When the TXRF spectrum of the solution containing 10 mg/L of Zn was measured, the SDD VITUS H7 was used. An X-ray tube 50 kV Magnum (Moxtek Inc., Orem, UT, USA) having a tantalum anode was operated at 25 kV and 0.2 mA. An X-ray waveguide, whose entrance and exit apertures had a height of 0.03 mm and a width of 1 cm, was placed between the X-ray tube and a sample holder where the dry residue of a sample solution was placed in order to obtain the collimated incident X-ray beam. An angle between the surface of a sample holder and the horizontal plane was set to 0.05°. All TXRF measurements were performed in air for 600 s.
The following equation12) was used to determine the detection limit for a target element in the dry residue of a sample solution:
| (1) |
Figure 1 shows a representative TXRF spectrum of the dry residue of a 1 μL droplet of the solution containing 10 mg/L of Zn measured with the SDD with an active area of 7 mm2 and that of the dry residue of a 1 μL droplet of the solution containing 5 mg/L of Zn measured with the SDD having an active area of 30 mm2. The Si Kα line, Ar K lines, and Ta L lines, which were due to the quartz glass substrates, air containing 0.9% of Ar, and the anode material of the X-ray tube, respectively, were observed in Fig. 1. The net intensity of the Zn Kα line in Fig. 1(a) was 15732 counts, and that in Fig. 1(b) was 20703 counts. Because 10 ng of Zn and 5 ng of Zn are calculated to be contained in the dry residue of a 1 μL droplet of the solution containing 10 mg/L of Zn and that of a 1 μL droplet of the solution containing 5 mg/L of Zn, respectively, the net intensity of the Zn Kα line per 1 ng of Zn in Fig. 1(b) was estimated to be about 2.6 times as high as that in Fig. 1(a). This result indicated that the net intensity of a fluorescent X-ray peak of an element per the mass of the element (hereinafter referred to as sensitivity) could increase with an increase in an active area of an SDD. The Fe Kα line in Fig. 1(b) and Ni Kα line in Figs. 1(a) and 1(b) would originate from components of the portable spectrometer, and the net intensities of these peaks in Fig. 1(b) were higher than those in Fig. 1(a). The enhancement in the net intensities of these peaks would be due to the use of the SDD with a larger active area. The spectral background in Fig. 1(b) was higher than that in Fig. 1(a). This was because the amount of the scattered X-rays reaching the X-ray detector increased when using the SDD with an active area of 30 mm2. Detection limits of Zn obtained from Figs. 1(a) and 1(b) were 87 pg and 65 pg, respectively. The use of the SDD with a 30 mm2 active area led to an improvement in the detection limit although the use of this detector resulted in an increase in the spectral background. Because the volume of the sample droplets was only 1 μL, detection limits expressed as the concentration of Zn were high. Detection limits obtained from Figs. 1(a) and 1(b) were 87 μg/L and 65 μg/L, respectively.
(a) Representative TXRF spectrum of the dry residue of a 1 μL droplet of a solution containing 10 mg/L of Zn measured with an SDD with an active area of 7 mm2 and (b) that of the dry residue of a 1 μL droplet of a solution containing 5 mg/L of Zn measured with an SDD with an active area of 30 mm2. The insets indicate enlarged spectra in the X-ray energy range from 5 to 8 keV.
Figure 2 shows representative TXRF spectra of the dry residues of 200 μL droplets of the solution containing 10 μg/L of Cr on a DLC sample holder and a hydrophobic film coated DLC sample holder and a representative TXRF spectrum of the dry residue of a 200 μL droplet of ultrapure water on a hydrophobic film coated DLC sample holder. These spectra were measured with the SDD having a 30 mm2 active area. The net intensity of the Cr Kα line in Fig. 2(a) was 11973 counts, and that in Fig. 2(b) was 24591 counts. The Cr peak was not observed in Fig. 2(c). The protrusion appeared on the high energy side of the Ar Kβ line as shown in Fig. 2(a), and this protrusion was also observed in Fig. 2(b). These protrusions were attributed to the K Kα line, and the K Kα line originated from the 1000 mg/L Cr standard solution used for preparing the solution containing 10 μg/L of Cr. The S, Cl, and Ca Kα lines would be due to contamination arising during the preparation of the sample dry residue. The dry residue of a 200 μL droplet of the solution containing 10 μg/L of Cr is calculated to contain 2 ng of Cr. A detection limit for Cr obtained from Fig. 2(b) was 26 pg, and that obtained from Fig. 2(a) was 50 pg. Preparing the dry residue of a large volume droplet of a very dilute aqueous solution sample made it possible to detect a low concentration of an element in the solution sample, and detection limits for Cr obtained from Figs. 2(a) and 2(b) were 0.25 μg/L and 0.13 μg/L, respectively. When a hydrophobic film coated DLC sample holder was used, the size of the dry residue of a 200 μL droplet of the solution containing 10 μg/L of Cr was reduced. A diameter of the dry residue of a 200 μL droplet of the solution containing 10 μg/L of Cr on a DLC sample holder was typically about 3 mm. On the other hand, a diameter of the dry residue of a 200 μL droplet of the solution containing 10 μg/L of Cr on a hydrophobic film coated DLC sample holder was typically several hundred μm. The use of a hydrophobic film coated DLC sample holder would lead to an enhancement in the ratio of fluorescent X-ray photons reaching the SDD to those emitted from the dry residue. Therefore, the net intensity of the Cr Kα line was enhanced when using this sample holder, and this enhancement led to an improvement in the detection limit for Cr.
Representative TXRF spectra of (a) the dry residue of a 200 μL droplet of a solution containing 10 μg/L of Cr on a DLC sample holder, (b) the dry residue of a 200 μL droplet of the solution containing 10 μg/L of Cr on a hydrophobic film coated DLC sample holder, and (c) the dry residue of a 200 μL droplet of ultrapure water on a hydrophobic film coated DLC sample holder. The insets indicate enlarged spectra in the X-ray energy range from 2 to 5 keV.
A previous paper11) presented TXRF spectra of the dry residues of 200 μL droplets of commercially available bottled drinking water and showed that a ring-shaped dry residue whose diameter of the ring was ten and several millimeters was formed when a 200 μL droplet of the commercially available bottled drinking water was dried on a DLC sample holder. On the other hand, a diameter of the dry residue of a 200 μL droplet of the solution containing 10 μg/L of Cr on a DLC sample holder was small as described above. The present study showed that the dry residue of a droplet of a very dilute aqueous solution could be small even when a few hundred μL of the solution is placed on a DLC sample holder. Because a high ratio of fluorescent X-ray photons reaching the SDD to those emitted from the dry residue would be obtained with the use of a DLC sample holder, a detection limit in the 100 ng/L range was achieved. As described above, the use of a hydrophobic film coated DLC sample holder reduced the size of the dry residue, leading to an improvement in the detection limit. In this study, a detection limit of one hundred and several tens of ng/L for Cr was achieved by using the portable TXRF spectrometer with a hydrophobic film coated DLC sample holder. This detection limit was about 0.02 times as high as the mean value of the concentration of Cr in an eluate obtained by soaking a stainless steel fork in 36 mL of ultrapure water at 60°C for 30 min, which was reported by Kawamura et al.,10) and it would be further improved by increasing the volume of a droplet of a sample solution placed on a sample holder. The present study indicated that measuring the dry residue of a sample droplet with large volume on a sample holder having a hydrophobic surface is effective for detecting low concentrations of elements in a high purity water sample. Therefore, the portable TXRF spectrometer with a hydrophobic film coated sample holder can be applied to on-site analysis of metals eluted from stainless steel products into high purity water for evaluating the corrosion resistance of them.
When the measurement of a very dilute aqueous solution sample was performed using a portable TXRF spectrometer, preparing the dry residue of a sample droplet with large volume on a hydrophobic film coated DLC sample holder was effective for improving the detection limit expressed as the concentration of a target element in the sample solution. This sample preparation method will be useful for trace elemental analysis of an eluate obtained by soaking a stainless steel product in high purity water using TXRF analysis, and it will make it possible to perform on-site evaluation of the corrosion resistance of stainless steel products using the portable TXRF spectrometer.
This study was supported by Kurita Water and Environment Foundation (No. 20A025).
