Highly efficient photochemical vapor generation of tellurium: effects of antimony and ferric ions

Abstract

In this work, a novel method for the highly sensitive detection of trace tellurium (Te) was developed based on photochemical vapor generation (PVG) coupled with inductively coupled plasma mass spectrometry (ICP-MS). Effects of antimony (Sb) and ferric ions (Fe) on the PVG of Te were reported for the first time. In a medium containing 2% (v/v) acetic acid (AA), 5.0 mg L⁻¹ of Sb(III), and 15.0 mg L⁻¹ of Fe(III), the conversion efficiency for both Te(IV) and Te(VI) was found to be 94% ± 3% upon 90 s of UV irradiation using a thin-film flow-through mercury lamp. The limit of detection (LOD, 3σ, n = 11) was 0.4 ng L⁻¹ for both Te(IV) and Te(VI), which was enhanced about 40 times for Te compared with that obtained using commercial pneumatic nebulization (PN) with ICP-MS. Good precision was achieved, with relative standard deviations (RSDs) of 2.1% and 2.5% for 1.0 µg L⁻¹ Te(IV) and Te(VI) standard solutions (n = 7), respectively. The method was successfully applied to determine trace inorganic Te in three environmental water samples with satisfactory results. Compared with the previous reported PVG systems for Te, the use of organic acids was significantly reduced by more than 10-fold while maintaining high sensitivity. The mechanism of PVG was investigated, and it was found that, in addition to volatile Te, volatile Sb and Sb nanoparticles were generated in this system. This study provides a new perspective for the direct analysis of the total amount of elements in environmental samples and contributes to understanding the interactions between Te, Sb, and Fe during photochemical processes.

---

Introduction

Photochemical vapor generation (PVG) employs radicals generated from the photodecomposition of low molecular weight organic acids (LMWOAs) to convert analytes into volatile species.^1–4 It serves as an alternative sample introduction technique in atomic spectrometry, characterized by its simplicity and efficiency.^5,6 PVG has been applied for the analysis of a wide range of elements, including traditional hydride-forming elements,^7,8 transition metals,^9,10 noble metals,^11,12 and halogens.^13–15 The analytical performance of these elements is fundamentally limited by their PVG efficiency. The use of nano-TiO₂ and metal–organic framework (MOF) materials effectively increases the PVG yields of analytes such as Te, Se, and As.^16–18 Moreover, the presence of transition-metal ions, such as Co, Ni, Fe, Cu, Cd, and V, in the PVG system can greatly enhance the PVG efficiency of most elements in the periodic table.^19–28 Synergistic effects between different transition metal ions have also been observed for the PVG of Mo, Re and Ru.^29–31

Due to its unique photoelectric and thermal properties, tellurium (Te) has been widely applied in industrial applications.³² Understanding the spatial and temporal distributions of tellurium in the environment provides a scientific basis for recognizing its migration and transformation patterns and assessing the impact of human activities on the environment.^33,34 In natural waters, tellurium predominantly exists as Te(IV) and Te(VI) at the ng L⁻¹ level.^35,36 Under UV irradiation, Te(IV) and Te(VI) can be converted into volatile forms in the presence of LMWOAs such as formic acid (FA) and acetic acid (AA).³⁷ However, their conversion efficiencies vary considerably. The highest efficiency for Te generation was achieved in a mixed FA and AA solution, followed by FA and AA used alone. Under those conditions, the PVG efficiency of Te(VI) is substantially lower than that of Te(IV).^8,38 Based on the differing PVG efficiencies of Te(IV) and Te(VI), speciation can be achieved. In a FA medium, for instance, Te(IV) can be converted into a volatile species under high-pressure mercury lamp irradiation (predominantly 365 nm irradiation) with a conversion efficiency of 20–30%, whereas Te(VI) is hardly transformed, enabling the selective determination of Te(IV). However, the addition of nano-TiO₂ to this system markedly enhanced the photochemical conversion of Te(VI), raising its efficiency to about 20%.³⁷ Furthermore, when Mn(II) and Fe(II) were introduced as sensitizers into a mixture of FA and AA and irradiated with a low-pressure mercury lamp (254 nm UV irradiation), the PVG efficiency of Te(IV) was greatly improved, while Te(VI) remained non-responsive, again allowing selective determination of Te(IV).³⁹ For the detection of total Te, pre-reduction of Te(VI) into Te(IV) was required. These pretreatment steps increase the risks of sample contamination and analyte loss, thereby compromising the accuracy of the analytical results. Recently, it was found that in a medium with 20% (v/v) AA and 2% (v/v) FA, the addition of Fe(II)/Fe(III) markedly enhanced the PVG efficiency of Te(IV) and Te(VI) using a thin-film flow-through mercury lamp (with 185 and 254 nm UV irradiation), but the signal response of Te(VI) was much lower than that of Te(IV). The addition of nano-TiO₂ in this system further largely improved the Te(IV) and Te(VI), making no difference in sensitivity between Te(IV) and Te(VI).⁴⁰ Likewise, in a homogeneous system with 9% (v/v) FA and 20% (v/v) AA, when V(V) was used as a mediator, no significant disparity in the analytical sensitivity between Te(IV) and Te(VI) was observed.⁴¹ These methods enable the direct detection of total Te. Nevertheless, they generally demand relatively high concentrations of organic acids, typically surpassing 20% (v/v), markedly elevating contamination risk and analytical cost, and necessitating substantial sample dilution.

In this study, an efficient PVG system was developed for the analysis of Te(IV) and Te(VI) in low concentrations of organic acids. The objective was to improve the conversion efficiency of tellurium while significantly reducing organic acid consumption. Inductively coupled plasma mass spectrometry (ICP-MS) was employed to investigate various influencing factors, including the type and concentration of mediators, the effects of mediators, the types and concentrations of LMWOAs, the irradiation time, and the wavelength. The developed method was applied for the detection of total Te in environmental samples, with successful results. The mechanism underlying this PVG process was also explored in this study.

Experimental section

Reagents

All reagents were of at least analytical reagent grade. Deionized water (DIW, 18 MΩ cm) used throughout the experiments was purchased from C'estbon (China Resources C'estbon Beverage Co., Ltd). Formic acid and acetic acid of ACS grade were purchased from Aladdin Industrial Corporation (Shanghai, China). Sodium tellurite and sodium tellurate dihydrate were dissolved in DIW to prepare 1000.0 mg L⁻¹ Te(IV) (Aladdin Industrial Corporation, Shanghai, China) and Te(VI) (Aladdin Industrial Corporation, Shanghai, China) stock solutions, respectively. Sb(III) standard solution prepared in DIW with a concentration of 1000.0 mg L⁻¹ was purchased from Beijing PUXI Standard Technology Corporation. Ferric chloride (Aladdin Industrial Corporation, Shanghai, China) was dissolved in DIW to prepare a 1000.0 mg L⁻¹ Fe(III) stock solution. The cation exchange resin used was purchased from Amberlite SR1L NA (Aladdin, Shanghai, China).

Instrumentation

Te detection was carried out using an ICP-MS instrument (LabMS 3000, Lab Tech, Beijing, China) as summarized in Table S1. Instead of the conventional pneumatic nebulization of ICP MS, the PVG system was adopted. This system was equipped with a 19 W thin-film flow-through mercury lamp (Beijing Titan Instruments Company, Beijing, China), which emitted UV radiation at wavelengths of 185 nm and 254 nm. A schematic diagram of the PVG system is presented in Fig. 1. The solution was introduced into the PVG reactor using an IFIS-D flow injector (Xi'an Remex Analysis Instrument Co., Ltd, Xi'an, China). To prevent excess water vapor and organic acid from entering the plasma, two gas–liquid separators (GLSs), each with an inner volume of 5.0 mL, were placed in series between the PVG unit and the ICP-MS. After photochemical reduction, the sample solution was transported into the GLSs, where the generated volatiles were separated from the solution and then directed to the ICP-MS for detection with the assistance of Ar carrier gas. Volatile species produced during the PVG were collected from the headspace of the first GLS in the PVG system. A 5.0 mL sample of the collected gas was injected and identified using gas chromatography-mass spectrometry (GC-MS) in split mode (10 : 1) on a Fuli-S900 GC-MSD instrument (Zhejiang, China), as summarized in Table S2. Nanoparticles produced via photochemical reduction in the liquid phase were characterized by X-ray photoelectron spectroscopy (XPS, AXIS Ultra DLD 800X, Kratos, UK) and by transmission electron microscopy coupled with energy-dispersive X-ray spectroscopy (TEM-EDS, FEI Tecnai G2 F20, FEI, USA). Details of the procedures used for sample preparation are provided in Text S1 and S2.

Fig. 1 Schematic of the PVG system.

Preparation of samples

River water samples were collected from Dongfeng canal, Jinjiang, and Funan river in Chengdu, Sichuan. All samples were filtered through 0.45 µm filters, and 1% (v/v) AA was added for preservation. The samples were then stored at 4 °C in the dark. The sample digestion of certified reference materials was performed according to previous reports.⁴¹ The digests were subsequently evaporated to near dryness at 100 °C to eliminate excess acids. Then, the digests were leached back and filled up to 10.0 mL with DIW. Cu was removed using a cation exchange resin. After that, 0.1 mL of the digest was transferred to the 10.0 mL centrifuge tube. The final samples contained 2% (v/v) AA, 5.0 mg L⁻¹ Sb(III), and 15.0 mg L⁻¹ Fe(III).

Analysis procedure

Firstly, 0.7 mL of sample solution was introduced into the PVG reactor (with an internal volume of 1.0 mL) using a flow injection pump operating at a flow rate of 6.0 mL min⁻¹. Subsequently, a blank sample solution was introduced into the system to push the sample solution into the photoreactor. The pump was then stopped for 80 s, resulting in a total UV irradiation time of 90 s. After that, the pump was reactivated to rapidly transport the sample solution through the GLSs. During this process, the volatile species produced were separated from the sample solution and subsequently transported to the ICP-MS for detection (Fig. 1). After each analysis, the PVG system was thoroughly rinsed with a 5% (v/v) H₂SO₄ solution, followed by deionized water (DIW), to remove any residual Te, Sb, or Fe. To clearly compare the differences in the Te signal response under various experimental conditions, the data were normalized. Specifically, the peak area obtained under the finally selected experimental conditions, including 2% (v/v) AA, 5.0 mg L⁻¹ Sb(III), and 15.0 mg L⁻¹ Fe(III) with 90 s irradiation (Fig. 2c), was set as 100%. The UV lamp remained on throughout the entire procedure.

Results and discussion

Effects of antimony and ferric ions on the PVG of Te

It was found that the presence of Sb exhibited an enhanced effect on the PVG of As.⁴² In this study, the effects of metalloid elements on the PVG of 1.0 µg L⁻¹ of Te(IV) and Te(VI) were evaluated in 2% (v/v) AA under 90 s UV irradiation, including Sb(III), As(III), Se(IV) and Bi(III). As illustrated in Fig. 2a, the photochemical reduction of Te(IV) and Te(VI) was greatly enhanced in all the investigated systems, with similar conversion efficiencies observed for Te(IV) and Te(VI). This confirms the enhancement effects of metalloid elements on the photochemical reduction of Te. Among them, maximum improvement was found in the Sb(III) containing system. Therefore, Sb(III) was selected in subsequent studies, and the effects of its concentration on the PVG of Te were carefully investigated. As shown in Fig. 2b, the signal responses for Te(IV) and Te(VI) were found to be rather low in the absence of Sb, and the analytical sensitivity of Te(VI) was significantly lower than that of Te(IV), which is consistent with previous reports.⁴⁰ As the Sb(III) concentration increased from 0 to 5.0 mg L⁻¹, significant improvement in the signal responses for both Te(IV) and Te(VI) was observed. The maximum signal responses for Te(IV) and Te(VI) were achieved at an Sb(III) concentration of 5.0 mg L⁻¹, and there was no obvious difference in sensitivity between Te(IV) and Te(VI) (Fig. S1). After that, the signal responses of Te(IV) and Te(VI) gradually decreased. Furthermore, the influence of Sb(V) on the signal responses of Te(IV) and Te(VI) was examined. The results indicated that similar effects were found with Sb(V) to those observed for the Sb(III) containing system. Considering the analytical sensitivity and reagent consumption, we selected 5.0 mg L⁻¹ Sb(III) for subsequent experiments.

Fig. 2 (a) Effects of metalloid concentration on 1.0 µg L⁻¹ of Te(IV)/Te(VI): 2% (v/v) AA and 90 s UV irradiation time; (b) effects of Sb(III)/Sb(V) concentration on 1.0 µg L⁻¹ of Te(IV)/Te(VI): 2% (v/v) AA and 90 s UV irradiation time; (c) effects of Sb(III) and transition metals on the detection of 1.0 µg L⁻¹ of Te(IV): 5 mg L⁻¹ Sb(III), 2% (v/v) AA, and 90 s UV irradiation time; (d) effects of Fe(III) and metalloid on the detection of 1.0 µg L⁻¹ of Te(IV)/Te(VI): 15 mg L⁻¹ Fe(III), 2% (v/v) AA, 10 mg L⁻¹ Sb(III)/As(III)/Se(IV)/Bi(III), and 90 s UV irradiation time.

Previous studies demonstrated the synergistic effect of Fe and nano-TiO₂ in the PVG of Te, where nano-TiO₂ promoted the matching of the conversion efficiency between Te(IV) and Te(VI), and Fe remarkably improved the signal response of Te.⁴⁰ Therefore, the effects of Fe(III), Cd(II), Co(II), and Ni(II) in the presence of Sb(III) were investigated in a system containing 2% (v/v) AA and 5.0 mg L⁻¹ Sb(III) under 90 s UV irradiation. As depicted in Fig. 2c, positive effects were observed in the systems containing Co(II) and Fe(III). Specifically, the addition of Fe(III) resulted in a significant 20% increase in the signal response of Te. Meanwhile, the presence of 5.0 mg L⁻¹ Cd(II) or Ni(II) had a slightly enhancing effect on the PVG of Te. However, higher concentrations of these ions led to a decrease in the signal responses of Te. Moreover, the effects of Fe(III) on the PVG of Te in the presence of other metalloid elements, including Se, As, and Bi, were also estimated in this study. The enhancement was also found in all of these systems, indicating the effects of Fe and metalloid elements on the photochemical reduction of Te (Fig. 2d). Consequently, the system containing Sb(III) and Fe(III) was selected for further studies, in consideration of the sensitivity.

The main factors influencing the PVG of Te and the detection of Te by ICP-MS were investigated in this study, including the type and concentration of LMWOAs, UV irradiation time, and carrier gas flow rate. The effects of FA and AA concentrations on the PVG signal responses of Te(IV) and Te(VI) with 5.0 mg L⁻¹ of Sb(III) were investigated, respectively. In the FA medium, the Te signal reached its maximum at 10–15% (v/v) FA; beyond this range, the response declined (Fig. 3a). In the AA-containing system, the signal responses of Te(IV) and Te(VI) increased almost linearly as AA concentration increased from 0 to 1% (v/v). The highest signal responses were observed in the range between 1% and 2%; after that, a decreased signal response was found (Fig. 3b). Compared to the system containing FA, the conversion of Te into volatile species was much more efficient in the AA medium. The effect of FA concentration on the PVG of Te in the presence of 2% (v/v) AA was also evaluated. Unlike previous reports,⁴¹ the addition of a low concentration of FA had no significant effect on the photochemical reduction of Te, while a high concentration of FA led to a decrease in the signal response of Te (Fig. 3c). Considering the signal response and reagent consumption, 2% (v/v) of AA was selected for subsequent studies.

Fig. 3 Effects of experimental conditions on the determination of 1.0 µg L⁻¹ Te(IV)/Te(VI): (a) concentration of AA: 5.0 mg L⁻¹ Sb(III), 15.0 mg L⁻¹ Fe(III), and 90 s irradiation time; (b) concentration of FA: 5.0 mg L⁻¹ Sb(III), 15.0 mg L⁻¹ Fe(III), and 90 s irradiation time; (c) concentration of mixed acid: 2% (v/v) AA, 5.0 mg L⁻¹ Sb(III), 15.0 mg L⁻¹ Fe(III), and 90 s irradiation time; and (d) irradiation time: 2% (v/v) AA, 5.0 mg L⁻¹ Sb(III), 15.0 mg L⁻¹ Fe(III).

UV irradiation plays a crucial role in the decomposition of organic compounds, the generation of radicals or intermediates, and the production of volatile elemental species. Thus, the impact of UV irradiation time, ranging from 0 to 120 s, on the photochemical reduction of Te(IV) and Te(VI) was studied in systems containing 2% (v/v) AA, 5.0 mg L⁻¹ Sb(III), and 15.0 mg L⁻¹ Fe(III). No signal response for Te(IV) and Te(VI) was observed when the UV lamp was off. The signal response increased almost linearly with increasing UV irradiation time from 20 to 80 s. Beyond 100 s, slight decreases in the signal response were noted (Fig. 3d). Obviously, the signal response of Te(VI) was lower than that obtained for Te(IV) with UV irradiation less than 90 s, and no significant difference in analytical sensitivity was observed between Te(IV) and Te(VI) above 90 s, suggesting relatively slow kinetics for volatile Te formation from Te(VI) compared to that of Te(IV) in the PVG system. The study also assessed the influence of UV irradiation wavelength on the PVG of Te(IV)/Te(VI). A quartz tube (45 cm × 4 mm outer diameter × 3 mm inner diameter) was employed as the reactor and placed close to a 15 W germicidal lamp emitting UV light at 254 nm or below. The sample volume used was the same as that employed in the thin-film UV lamp (with irradiation at 185 and 254 nm). As illustrated in Fig. S2, the maximum signal response for Te observed at 50 min UV irradiation was approximately 37% of that obtained using the thin-film UV lamp. This suggested relatively slow kinetics for the photochemical reduction of Te under such conditions, and vacuum UV irradiation was necessary for the highly efficient reduction of Te. The optimum flow rates for Ar1 and Ar2 (as shown in Fig. 1) were 0.3 L min⁻¹ and 0.6 L min⁻¹.

Interference study

Coexisting ions may interfere with the detection of analytes in the PVG system. Therefore, the effects of coexisting ions on the determination of 1.0 µg L⁻¹ Te(IV) and Te(VI) were evaluated in the presence of 2% (v/v) AA, 5.0 mg L⁻¹ Sb(III), and 15.0 mg L⁻¹ Fe(III). As evident from Table 1, 5.0 mg L⁻¹ of K(I), Ca(II), Na(I), Mg(II), Cd(II) and V(V) had no significant effect on the detection of 1.0 µg L⁻¹ Te. Also, 10.0 mg L⁻¹ of Mn(II), Cr(III), and Ni(II) had no significant effect on the detection of 1.0 µg L⁻¹ Te. The effects of 2.0 mg L⁻¹ of Co(II), Se(VI), and 0.5 mg L⁻¹ of Se(IV) were not significant. The effect of adding additional Fe(III) and Sb on the Te response in this system was investigated; the results showed that the addition of an additional 5 mg L⁻¹ Sb(III) and 50 mg L⁻¹ Fe(III) caused no significant effect on Te determination under the studied conditions. Additionally, the impacts of common anions and inorganic acids on the determination of Te were also investigated. The influences of 10.0 mg L⁻¹ SO₄²⁻ and Cl⁻, 20.0 mg L⁻¹ NO₃⁻, 0.05% (v/v) H₂SO₄, 0.02% (v/v) HCl and 0.01% (v/v) HNO₃ were not notable; above these concentrations, changes of more than 10% were observed in the Te signal responses.

Table 1 Effect of interferences in the detection of 1.0 µg L⁻¹ Te

Ions/acids	Used reagents	Concentration/mg L^−1	[Ions]/[Te]	Te(IV) recovery/% (n = 3)	Te(VI) recovery/% (n = 3)
a Recovery of Te obtained after using 0.2 g mL^−1 of cation exchange resin for Cu removal.
K(I)	KCl	5.0	5000	98 ± 2	97 ± 1
Ca(II)	CaCl2	5.0	5000	99 ± 1	97 ± 2
Na(I)	NaCl	5.0	5000	97 ± 2	98 ± 2
Mg(II)	MgCl2	5.0	5000	103 ± 3	106 ± 3
Cd(II)	CdCl2	5.0	5000	108 ± 3	109 ± 3
V(V)	NH4VO3	5.0	5000	97 ± 4	97 ± 2
Mn(II)	MnSO4	10.0	10 000	90 ± 2	97 ± 4
Cr(III)	CrCl3	10.0	10 000	96 ± 2	95 ± 2
Ni(II)	NiSO4	10.0	10 000	94 ± 3	93 ± 2
Co(II)	CoSO4	2.0	2000	105 ± 3	104 ± 2
Se(VI)	Na2SeO4	2.0	2000	103 ± 2	105 ± 2
Se(IV)	Na2SeO3	0.5	500	107 ± 3	106 ± 2
Cu(II)	CuSO4	0.02	20	92 ± 3	92 ± 2
		0.05	50	72 ± 2	63 ± 1
		1.0^a	1000	93 ± 2	92 ± 1
Sb(III)	Standard solution	5.0	5000	98 ± 1	97 ± 2
Fe(III)	FeCl3	50.0	50 000	93 ± 2	92 ± 1
SO4^2−	Na2SO4	10.0	10 000	98 ± 2	93 ± 1
NO3^−	KNO3	20.0	20 000	93 ± 3	94 ± 1
Cl^−	NaCl	10.0	10 000	95 ± 2	98 ± 3
H2SO4		0.05% (v/v)	—	92 ± 2	93 ± 3
		0.1% (v/v)		79 ± 1	78 ± 3
HCl		0.02% (v/v)	—	93 ± 2	92 ± 2
		0.05% (v/v)		81 ± 1	80 ± 2
HNO3		0.01% (v/v)	—	92 ± 2	94 ± 2
		0.02% (v/v)		85 ± 2	87 ± 1

---

According to previous reports, the influence of Cu(II) was severe for Te detection, either in HG or PVG.^41,43 The effect of Cu(II) concentration on Te determination was therefore studied in this work. The addition of 0.02 mg L⁻¹ of Cu(II) did not affect the measurement of 1.0 µg L⁻¹ Te. However, an increase in Cu(II) concentration resulted in a reduced signal response for Te, and the recovery of 1.0 µg L⁻¹ Te after the addition of 0.05 mg L⁻¹ of Cu(II) was only 72%; this was attributed to the formation of Te–Cu colloids. Therefore, the developed method can allow direct detection of Te in natural waters, as the concentration of Cu is normally lower than 20 µg L⁻¹. However, for samples with higher concentrations of Cu, the removal Cu from the sample solution prior to analysis is necessary. The use of a 0.2 g mL⁻¹ cation exchange resin has been demonstrated to be an effective approach to alleviate the effect of Cu. The treatment was as follows: the cation exchange resin was added to a solution containing Cu(II) and 1.0 µg L⁻¹ Te. The solution was manually shaken for one minute and then allowed to stand for 30 s to enable the resin to settle to the bottom of the vessel. Subsequently, the supernatant was analyzed by adding the requisite amounts of AA, Sb(III), and Fe(III). It was found that the presence of Cu(II) at a concentration of up to 1 mg L⁻¹ did not significantly interfere with the detection of Te, with a recovery rate of approximately 93%. However, for samples with an unknown matrix, it is recommended to remove Cu prior to analysis to ensure accurate results.

Mechanism of the proposed PVG system

The mechanism of Sb and Fe-assisted PVG of Te was studied. The UV-visible absorption spectra showed that the presence of Sb and Fe promoted the absorption of UV irradiation in the sample solution, which may be beneficial for the photochemical reduction of Te, based on the first law of photochemical reactions (Fig. S4).⁴⁴ The volatile species of Te generated in the photochemical reaction were identified by GC-MS. The major fragments in the mass spectra were recorded as Te⁺, [(CH₃)Te]⁺, and [(CH₃)₂Te]⁺, corresponding to m/z 130, 145, and 160, which implied the formation of (CH₃)₂Te from Te(IV) and Te(VI) after photochemical reduction, as shown in Fig. 4a and b. Additionally, (CH₃)₃Sb was detected along with (CH₃)₂Te (Fig. S5). However, no volatile Fe species were detected by GC-MS and ICP-MS, indicating that the PVG medium was unfavorable for the generation of volatile Fe species such as Fe(CO)₅.

Fig. 4 (a and b) Mass spectra of volatile species generated from photochemical reduction of 5.0 mg L⁻¹ Te(IV) and Te(VI), 5.0 mg L⁻¹ Sb(III), 15.0 mg L⁻¹ Fe(III), 2% (v/v) AA, and 90 s irradiation time. (c and d) XPS spectra of the liquid phase products after UV irradiation: 10.0 mg L⁻¹ Te, 15.0 mg L⁻¹ Sb(III), 40.0 mg L⁻¹ Fe(III), 10% (v/v) AA, and 90 s irradiation time.

To figure out the photochemical process of Te and Sb in the liquid phase, a solution containing 10.0 mg L⁻¹ Te, 5.0 mg L⁻¹ Sb(III), 15.0 mg L⁻¹ Fe(III), and 2% (v/v) AA was characterized by TEM after 90 s of irradiation. Dispersed or aggregated nanoparticles with diameters below 100 nm were found, as shown in Fig. 5 and S6. Bimetallic nanoparticles of Te and Sb were found in the liquid solution, and dispersed nanoparticles of Fe were also observed by EDS analysis. XPS was used to analyze the valence states of elements in nanoparticles generated in the liquid. Due to the low sensitivity of XPS, a solution containing 10.0 mg L⁻¹ Te, 15.0 mg L⁻¹ Sb(III), 40.0 mg L⁻¹ Fe(III), and 10% (v/v) AA was used for analysis. XPS peaks were assigned according to the NIST XPS database. It was found that Te existed mainly as Te(0) and Te(IV), with two peaks at binding energies of 572.7 eV and 583.0 eV belonging to the ³D_5/2 and ³D_3/2 of Te(0), and two peaks at binding energies of 576.0 eV and 586.5 eV belonging to the ³D_5/2 and ³D_3/2 of Te(IV) (Fig. 4c). Furthermore, two peaks at binding energies of 528.2 eV and 537.6 eV belonged to the ³D_5/2 and ³D_3/2 of Sb(0), and two peaks at binding energies of 530.2 eV and 539.4 eV belonged to the ³D_5/2 and ³D_3/2 of Sb(III), indicating that Sb existed predominantly as Sb(III) and Sb(0) in the nanoparticles (Fig. 4d).

Fig. 5 (a and b) TEM images and (c–g) EDS characterization of nanoparticles in the liquid phase (with Te): 10.0 mg L⁻¹ Te, 5.0 mg L⁻¹ Sb(III), 15.0 mg L⁻¹ Fe(III), 2% (v/v) AA, and 90 s irradiation time.

Based on the above investigations, the PVG process of Te was speculated to proceed as follows: AA is decomposed under UV irradiation to form solvated electrons (e_aq⁻) and reductive radicals such as ˙H, ˙CH₃, and ˙CO₂⁻, which react with Te(IV) and Te(VI) to generate Te⁰.^6,45 Subsequently, Te⁰ is attacked by ˙CH₃ to form (CH₃)₂Te. The presence of Fe accelerates the decomposition of AA and facilitates the formation of reductive radicals.²² The addition of Sb leads to a decreased amount of ˙OH in the liquid and an increased ratio of reducing radicals to oxidizing radicals, based on previous reports, thus promoting the photochemical reduction of Te.⁴² Nanoparticles of Sb and Fe generated during the photochemical process may attract radicals onto their surface and provide a highly active area for Te reduction, which can be supported by the results of the formation of bimetallic nanoparticles of Te and Sb in solution as described in previous reports.^6,46 However, this needs further investigation and is beyond the scope of this study.

Analytical performance and sample analysis

The analytical performance of the developed system was evaluated in the medium of 2% (v/v) AA, 5.0 mg L⁻¹ Sb(III), and 15.0 mg L⁻¹ Fe(III) with 90 s irradiation. Standard solutions of Te(IV) and Te(VI) were prepared to establish the calibration curves, respectively. There was no significant difference in analytical sensitivity for Te(IV) and Te(VI) (Fig. S8). The linear regression coefficients (R²) were 0.998 for Te(IV) and 0.997 for Te(VI). The precision expressed as relative standard deviations (RSDs) for seven replicate measurements of 1.0 µg L⁻¹ of Te(IV)/Te(VI) standard solutions was 2.1% and 2.5%, respectively. The method limit of detection (LOD, 3σ) was calculated to be 0.4 ng L⁻¹ for Te(IV)/Te(VI) when collision mode was used, which was improved about 40-fold compared to that using direct sample introduction solution nebulization. Considering the sample dilution prior to analysis (usually one-fold), the LOD was enhanced about 20-fold. The conversion efficiency was determined at a Te concentration of 10.0 µg L⁻¹ by comparing the relative concentrations of Te in solutions with and without UV irradiation. The concentration was measured by PN-ICP-MS after evaporating the solutions to near dryness and re-dissolving them in 2% (v/v) HNO₃. The conversion efficiencies of Te(IV) and Te(VI) were 95% ± 2% and 94% ± 3%, respectively, which are comparable to those obtained in the PVG system containing Fe(II) and nano-TiO₂, with a conversion efficiency of about 91% ± 2%. A comparison of the analytical performance of the developed method to those in previous reports is presented in Table S3. The proposed method was sensitive, and the reagent consumption could be 10-fold less than that of other PVG systems, leading to improved LOD for Te.

The method was employed for the analysis of three environmental water samples using external calibration, and the results are shown in Table 2. The recovery was found to be in the range of 90% and 105%, demonstrating the feasibility of the method for the detection of trace Te in natural water samples. A spiked test was carried out to further evaluate the accuracy of the proposed method; soil and sediment certified reference materials (GBW07407a, GBW07305a) were analyzed. A standard addition method was applied to assess the interference from the sample matrix for Te detection. There was no significant difference between the measured values obtained by this method and the certified results, as shown in Table 3.

Table 2 Results for the detection of Te in natural water samples

Sample	Measured^a (µg L^−1)	Added (µg L^−1)	Found^a (µg L^−1)	Recovery (%)
a Mean value ± standard deviation (n = 3).
Funan river	0.024 ± 0.002	0.020	0.042 ± 0.001	91
Jinjiang river	0.018 ± 0.002	0.020	0.038 ± 0.002	100
Dongfeng river	0.022 ± 0.001	0.020	0.043 ± 0.002	105

---

Table 3 Analytical results of Te in CRMs

Sample	Measured (µg g^−1, n = 3)	Reference value (µg g^−1)
GBW07407a	0.059 ± 0.002	0.06
GBW07305a	0.29 ± 0.02	0.3

---

Conclusion

In this study, an efficient PVG reaction system for the determination of total Te was successfully established. For the first time, the effects of metalloid elements and Fe were studied for the PVG of Te. Under the selected conditions, Te(IV) and Te(VI) can be equally and efficiently converted into volatile species, making the direct detection of total Te possible. The analytical sensitivity of the developed method was increased by about 40-fold compared to the PN method. Furthermore, the reagent consumption of organic acids in this study can be reduced more than 10-fold in comparison with previously reported PVG systems, resulting in greatly reduced contamination risk for trace-element detection. However, it should be noted that Sb(III) is toxic and a suspected carcinogen, and careful considerations for disposal are required to prevent harmful effects on the environment. In addition to Te, methods mediated by metalloids can be employed to determine the total concentrations of elements such as As, Se, and Bi. (CH₃)₂Te, (CH₃)₃Sb, and nanoparticles of Sb were formed after UV irradiation in this study, which provided important scientific evidence for understanding the photochemical behavior of metalloid elements in environments, especially in areas with antimony mines with high levels of Sb and Fe.

Author contributions

Weiwen Huang: conceptualization, methodology, data curation, and writing; Ying Yu: investigation and writing – review; Liang Dong: investigation and writing – review; Xiuqin Deng: investigation and writing – review; Xinyi Zhao: investigation and writing – review; Liwei Liu: investigation and writing – review; Ying Gao: supervision.

Conflicts of interest

There are no conflicts to declare.

Data availability

Data supporting the experimental findings of this study are presented within the article; those mentioned as a simple summary statement are available from the corresponding author upon reasonable request.

Supplementary information: additional information, as noted in the text, includes signal responses of Te(IV) and Te(VI), UV irradiation time using a germicidal lamp, effect of Ar carrier gas flow rate, UV-vis absorbance of the PVG medium, mass spectra of volatile Sb species, characterization by TEM and XPS, and the comparison of LODs. See DOI: https://doi.org/10.1039/d5ja00268k.

Acknowledgements

This work was financially supported by the National Natural Science Foundation of China (no. 41973019) and the Scientific Research Fund of the Science and Technology Department of Sichuan Province (no. 2023JDRC0005).

References

1. X. M. Guo, R. E. Sturgeon, Z. Mester and G. J. Gardner, Anal. Chem., 2004, 76, 2401–2405.
2. X. Guo, R. E. Sturgeon, Z. Mester and G. J. Gardner, Anal. Chem., 2003, 75, 2092–2099.
3. Q. Wang, J. Liang, J. Qiu and B. Huang, J. Anal. At. Spectrom., 2004, 19, 715–716.
4. Y. Gao, X. Peng, Z. Shi, R. Zhang, X. Xia, F. Yue and R. Liu, At. Spectrosc., 2012, 33, 73–77.
5. D. Leonori and R. E. Sturgeon, J. Anal. At. Spectrom., 2019, 34, 636–654.
6. R. E. Sturgeon, E. Pagliano, G. S. Lopes, R. S. Neto and J. K. Brito, J. Anal. At. Spectrom., 2025, 40, 70–97.
7. X. Deng, J. Liu, K. Yuan, Y. Huang, L. Dong, Z. Shi and Y. Gao, Anal. Chem., 2024, 96, 18337–18342.
8. K. A. Romanovskiy, M. A. Bolshov, A. V. Münz, Z. A. Temerdashev, M. Y. Burylin and K. A. Sirota, Talanta, 2018, 187, 370–378.
9. Y. Gao, R. E. Sturgeon, Z. Mester, E. Pagliano, R. Galea, P. Saull, X. Hou and L. Yang, Microchem. J., 2016, 124, 344–349.
10. C. Zheng, R. E. Sturgeon, C. S. Brophy, S. He and X. Hou, Anal. Chem., 2010, 82, 2996–3001.
11. R. M. de Oliveira and D. L. Borges, J. Anal. At. Spectrom., 2018, 33, 1700–1706.
12. Q. Yang, H. Chen, J. Hu, K. Huang and X. Hou, Anal. Chem., 2021, 94, 593–599.
13. J. Hu, R. E. Sturgeon, K. Nadeau, X. Hou, C. Zheng and L. Yang, Anal. Chem., 2018, 90, 4112–4118.
14. R. E. Sturgeon, Anal. Chem., 2015, 87, 3072–3079.
15. P. Grinberg and R. E. Sturgeon, J. Anal. At. Spectrom., 2009, 24, 508–514.
16. H. Li, Y. Luo, Z. Li, L. Yang and Q. Wang, Anal. Chem., 2012, 84, 2974–2981.
17. J. Jia, Z. Long, C. Zheng, X. Wu and X. Hou, J. Anal. At. Spectrom., 2015, 30, 339–342.
18. C.-H. Lin, Y. Chen, Y.-A. Su, Y.-T. Luo, T.-T. Shih and Y.-C. Sun, Anal. Chem., 2017, 89, 5891–5899.
19. W. Zeng, J. Hu, H. Chen, Z. Zou, X. Hou and X. Jiang, J. Anal. At. Spectrom., 2020, 35, 1405–1411.
20. Q. Mou, L. Dong, L. Xu, Z. Song, Y. Yu, E. Wang, Y. Zhao and Y. Gao, J. Anal. At. Spectrom., 2021, 36, 1422–1430.
21. J. Mo, Q. Li, X. Guo, G. Zhang and Z. Wang, Anal. Chem., 2017, 89, 10353–10360.
22. Y. Yu, Y. Jia, Z. Shi, Y. Chen, S. Ni, R. Wang, Y. Tang and Y. Gao, Anal. Chem., 2018, 90, 13557–13563.
23. R. M. de Oliveira, D. L. Borges, P. Grinberg and R. E. Sturgeon, J. Anal. At. Spectrom., 2021, 36, 2097–2106.
24. R. M. de Oliveira, D. L. Borges, P. Grinberg, Z. Mester and R. E. Sturgeon, J. Anal. At. Spectrom., 2021, 36, 1235–1243.
25. R. E. Sturgeon and E. Pagliano, J. Anal. At. Spectrom., 2020, 35, 1720–1726.
26. F. Xu, Z. Zou, J. He, M. Li, K. Xu and X. Hou, Chem. Commun., 2018, 54, 4874–4877.
27. J. Vyhnanovský, R. E. Sturgeon and S. Musil, Anal. Chem., 2019, 91, 13306–13312.
28. L. Dong, Y. Ning, J. Hu, W. Wang, Y. Yu and Y. Gao, J. Anal. At. Spectrom., 2024, 39, 791–799.
29. J. Hu, H. Chen, X. Hou and X. Jiang, Anal. Chem., 2019, 91, 5938–5944.
30. S. Musil, J. Vyhnanovský and R. E. Sturgeon, Anal. Chem., 2021, 93, 16543–16551.
31. Y. Zhen, H. Chen, M. Zhang, J. Hu and X. Hou, Appl. Spectrosc. Rev., 2022, 57, 318–337.
32. A. Qvarforth, A. Augustsson, M. von Ehr, G. Mandava, I. Rodushkin, E. Engstrom, S. Eisele and J. Lundqvist, Environ. Sci. Technol., 2025, 59, 1145–1155.
33. M. Filella, C. Reimann, M. Biver, I. Rodushkin and K. Rodushkina, Environ. Chem., 2019, 16, 215–228.
34. J. A. Wiklund, J. L. Kirk, D. C. Muir, J. Carrier, A. Gleason, F. Yang, M. Evans and J. Keating, Environ. Sci. Technol., 2018, 52, 6137–6145.
35. N. Belzile and Y.-W. Chen, Appl. Geochem., 2015, 63, 83–92.
36. D. Spanu, C. Omodei, G. Binda, T. Grande and S. Recchia, J. Anal. Test., 2025, 1–13.
37. C. Zheng, Q. Ma, L. Wu, X. Hou and R. E. Sturgeon, Microchem. J., 2010, 95, 32–37.
38. D. P. de Quadros and D. L. Borges, Microchem. J., 2014, 116, 244–248.
39. E. Jeníková, E. Nováková, J. Hraníček and S. Musil, Anal. Chim. Acta, 2022, 1201, 339634.
40. H. He, X. Peng, Y. Yu, Z. Shi, M. Xu, S. Ni and Y. Gao, Anal. Chem., 2018, 90, 5737–5743.
41. L. Dong, H. Chen, Y. Ning, Y. He, Y. Yu and Y. Gao, Anal. Chem., 2022, 94, 4770–4778.
42. X. Deng, L. Dong, H. Chen, W. Wang, Y. Yu and Y. Gao, Anal. Chem., 2023, 96, 652–660.
43. Y.-W. Chen, A. Alzahrani, T.-L. Deng and N. Belzile, Anal. Chim. Acta, 2016, 905, 42–50.
44. R. E. Sturgeon and P. Grinberg, J. Anal. At. Spectrom., 2012, 27, 222–231.
45. R. E. Sturgeon, in Vapor Generation Techniques for Trace Element Analysis, Elsevier, 2022, pp. 213–263.
46. R. Bar-Ziv, I. Zilbermann, O. Oster-Golberg, T. Zidki, G. Yardeni, H. Cohen and D. Meyerstein, Chem.–Eur. J., 2012, 18, 4699–4705.

---