Low cost sample preparation method using ultrasound for the determination of environmentally critical elements in seaweed

Graphical abstract

Highlights

• •Extraction of environmentally critical elements from seaweed was performed by UAE.
• •The optimized UAE method was simple, employed diluted reagents and short times.
• •Only when US was used the extraction efficiency was satisfactory for all analytes.
• •The developed UAE method is accurate and presents relatively low LOQs.
• •Environmentally critical elements were determined in seaweed from the Antarctic region.

Abstract

In this study, ultrasound (US) was evaluated for As, Cd, Pb, Mn, Sr and V extraction from seaweed samples. The following parameters of ultrasound-assisted extraction (UAE) using an US bath were: frequency (25 to 130 kHz), amplitude (30 to 100%), temperature (30 to 80 °C), sample mass (50 to 200 mg), extractant concentration (1 to 3 mol L⁻¹ of HNO₃) and treatment time (5 to 30 min). Acoustic density and power density distribution were calculated using the calorimetric method and mapping of the acoustic pressure distribution was also evaluated. The optimized UAE conditions were 200 mg of sample in 10 mL of 2 mol L⁻¹ HNO₃ and 30 min of sonication in a 25 kHz US bath (37.2 ± 4.0 W L⁻¹) at 70% of amplitude and 70 °C. Analytes were quantified using inductively coupled plasma mass spectrometry and results were compared with values obtained using “silent” conditions (magnetic or mechanical stirring at 500 rpm, and without stirring), and a reference method based on microwave-assisted wet digestion (MAWD). The UAE method demonstrated the best extraction efficiency (higher than 95%) for all analytes, especially for As, Cd and V, with lower standard deviations (up to 5%) and lower blank values in comparison with the silent conditions. The proposed UAE method was more advantageous than the reference method, being faster, simpler, safer, more environmentally friendly, and with higher detectability (lower limits of quantification, from 0.0033 to 1.34 µg g⁻¹). In addition, negligible blank values were obtained for UAE and no interference were observed in the determination step. Furthermore, the optimized UAE method was applied for Antarctic seaweed samples and comparison with results obtained by MAWD was satisfactory. In this sense, UAE is demonstrated to be a suitable option for sample preparation of seaweed samples and further determination of environmentally critical elements avoiding the use of concentrated reagents as in the MAWD reference method.

1. Introduction

The presence of environmentally critical elements (such as As, Cd, Pb, Mn, Sr, and V) in marine environment can be derived from natural sources, as well as contamination from industrial, agriculture and municipal waste transport. Bioaccumulation of these elements can happen in animals and aquatic plants, potentially causing environmental damage [1], [2]. Moreover, the presence of As, Cd, Pb and V in the marine environment may cause the contamination of other food sources (such as seafood). Thus, monitoring these elements is essential in marine environments. In this sense, environmental bioindicators such as macro algae, can be used [3]. Macro algae are ubiquitous aquatic plants in almost all marine ecosystems, having been used as bioindicators of As, Cd, Pb, Mn, Sr, and V due to the presence of polysaccharides and proteins in their structure that have a high affinity to these elements [4].

The determination of environmentally critical elements can be performed using plasma-based techniques (inductively coupled plasma optical spectrometry, ICP-OES, and inductively coupled plasma mass spectrometry, ICP-MS) [5], [6]. Furthermore, ICP-MS is the recommended technique for element determination in seaweed, due to low limits of quantification (LOQs), in the range of parts per trillion, its wide calibration range, its high precision and its multi-element capacity [7]. However, with conventional sample introduction modes, the ICP-MS equipment requires samples to be in a solution form. In this sense, a previous sample preparation step is mandatory [4], [8], [9], [10], [11].

Conventional sample preparation methods for decomposition of organic samples aiming at element determination, such as microwave-assisted wet digestion (MAWD), generally employ relatively long processing times and the use of concentrated acids, resulting in high reagent consumption and safety risks for the operator. Furthermore, the equipment to perform MAWD procedures is usually relatively expensive [12].

On the other hand, on the quest of following green chemistry principles, the reduction of time, cost, energy, reagents and waste generation are a necessity when developing new sample preparation methods [13], [14]. In this way, ultrasound (US) is a promising alternative for sample preparation, as it can promote a reduction of process time, temperature, use of reagents and waste generation [15].

When US is applied to liquid medium, successive cycles of rarefaction and compression promote the formation, growth and implosion of acoustic cavitation bubbles [16]. When these cavitation bubbles implode, “hot spots” with high temperatures and pressures (around 5000 K and 1700 atm), as well as shock waves, are formed [10], [17]. Besides, asymmetric implosion occurs when the bubbles implode near an interface, which generates high speed liquid microjets (∼400 km h⁻¹) [18]. The microjets, shock waves and high localized pressures and temperatures can erode the surface of solid samples, which can be highly beneficial to sample preparation methods based on extraction processes [19], [20].

Some effects of US, such as shear forces, heating, acoustic cavitation and microjets can be explored for ultrasound-assisted extraction (UAE), and can enhance extraction efficiencies when compared to more widely used extraction methods [21], [22], [23]. In this sense, UAE has been widely applied for extraction of bioactive compounds from seaweed, such as proteins [24], phenolic compounds [25], [26] and pigments [27]. However, this sample preparation method has yet to be used for environmentally critical elements extraction from seaweed.

For this reason, this study evaluates, for the first time, the use of UAE as a sample preparation method for further determination of As, Cd, Mn, Pb, Sr and V in seaweed by ICP-MS looking for a lower consumption of reagents. The US frequency, amplitude, temperature, extractant concentration, sonication time and sample mass were evaluated. The proposed method was applied to commercial and Antarctic seaweed samples. For accuracy evaluation, the results of the UAE method were compared with those obtained by a reference method based on MAWD. Furthermore, analyte addition studies and certified reference material (CRM) analysis (aquatic plant, BCR 60) were carried out. Moreover, the influence of US was evaluated and different approaches and US frequencies were evaluated, as well as the comparison with silent conditions. Moreover, acoustic density and power density distribution were calculated using the calorimetric method and mapping of the acoustic pressure distribution was also evaluated.

2. Materials and methods

2.1. Instrumentation

For the MAWD procedure (reference method), a Multiwave 3000 microwave system (Anton Paar, Austria) equipped with 16 modified polytetrafluoroethylene (PTFE-TFM) vessels was used (maximum pressure of 40 bar and maximum temperature of 220 °C).

For the UAE procedures, three bath-type systems were evaluated: system A, with an input power of 100 W, operating at 25 or 45 kHz (Sonic TI-H 5 3.5 L, Elma GmbH & Co., Germany); system B, with an input power of 200 W, operating at 35 or 130 kHz (Sonic TI-H-10 8.6 L, Elma GmbH & Co.); and system C, with an input power of 330 W, operating at 37 or 80 kHz (Sonic P120H 9.0 L, Elma GmbH & Co.).

A high intensity mechanical stirrer (homogenizer tip diameter of 10 mm, PT3100 D, Polytron, Switzerland) was used as silent condition, operating at 500 rpm. A magnetic stirring system was also used as silent condition, using a hotplate stirring system (AREX-F20500413, Velp Scientifica srl, Italy).

The determination of As, Cd, Mn, Pb, Sr and V was performed using a NexION 300x inductively coupled plasma mass spectrometer (Perkin Elmer, Canada), assembled in accordance to a previous study [28]. A Spectro Ciros CCD inductively coupled plasma optical emission spectrometer (Spectro Analytical Instruments, Germany) was used for determination of carbon content in the extracts. The operational conditions of both spectrometers are described in Table S1 (Supplementary Material).

For the morphology and structure characterization of seaweed samples (before and after US treatment) a Sigma 300 VP scanning electron microscope (Carl Zeiss, Germany) was used, assembled according to a previous study [29]. Particle size distribution before and after UAE procedures and silent conditions was determined by laser diffraction technique using a Mastersizer 2000 system (Malvern Instruments, United Kingdom).

2.2. Reagents

Purified water (18.2 MΩ cm, purification system from Millipore, USA) was used to prepare all solutions. Only analytical grade reagents were used. Distilled HNO₃ (65%, Merck) was obtained using a sub-boiling system (duoPUR, Milestone, Italy).

A 10 mg L⁻¹ multielement stock solution (SCP33MS, SCP Science, Canada) was used to prepare analytical standards by sequential dilution in 5% HNO₃ (v v⁻¹) for As, Cd, Pb, Mn, Sr and V determination by ICP-MS. Citric acid (Synth, Brazil) was used to prepare a 10000 mg L⁻¹ stock solution for C determination by ICP-OES. For C determination, Ar at 0.1 L min⁻¹ was used to purge samples and standards for 2 min prior determination, and Y (final concentration of 1 mg L⁻¹ in samples and standards, Spex Cert Prep, USA) was used as internal standard. Argon (99.998% of purity, White Martins, Brazil) was used for both ICP-MS and ICP-OES determinations.

2.3. Samples and certified reference material

Six seaweed samples were used. Commercial edible Nori seaweed (Porphyna spp) from Korea (sample A) was used for optimization of UAE parameters. The optimized UAE method was then applied to different seaweed species from the Antarctic region, namely Iridaea cordata (sample B), Curdiea racovitzae (sample C), Desmarestia anceps (sample D), Himantothallus grandifolius (sample E), and Desmarestia antarctica (sample F). The samples were dried and ground in an impact grinding mill (model A 11 basic, IKA®, Germany). For accuracy evaluation, a CRM of aquatic plant, BCR60 (Community Bureau of Reference, European Commission), was used.

2.4. Methods

2.4.1. Microwave-assisted wet digestion of seaweed samples

The reference method (MAWD) was carried out using 300 mg of seaweed sample and concentrated HNO₃ as the digesting solution (6 mL). The following irradiation program was used, composed of three steps: i) 20 min to 700 W; ii) 20 min at 700 W, and iii) cooling to 50 °C (25 min). After decomposition, the solution was made up to 25 mL with water in a volumetric flask for further As, Cd, Mn, Pb, Sr and V determination by ICP-MS and C determination by ICP-OES.

2.4.2. Ultrasound-assisted extraction parameters

The UAE procedures were performed in 15 mL polypropylene vessels with 1.5 cm of internal diameter, 11.7 cm of height and conical bottom. The samples were weighed directly in the vessels, the extractant was added and sonication was carried out.

In order to assure reproducible conditions, the vessels were positioned at a depth of 8.5 cm in the positions with higher acoustic field intensity of the US baths. The acoustic intensity was determined in accordance to a previous study [30].

The extracts were centrifuged (centrifuge model Q-222 T208, Quimis, Brazil) for 15 min at 4000 rpm, and filtered (cellulose filters, ɸ 0.45 µm, CA, Chamafil®Xtra, Germany). After filtration, As, Cd, Mn, Pb, Sr and V were determined by ICP-MS. For the optimized UAE conditions, dissolved C was also determined by ICP-OES.

The starting parameters of the UAE procedures were 50 mg of sample and 10 mL of extractant (1 mol L⁻¹ HNO₃). Initially, different bath systems were evaluated, namely baths A (25 and 45 kHz, 100 W), B (35 and 130 kHz, 200 W) and C (37 and 80 kHz, 330 W).

Output power and acoustic density were calculated using the calorimetric method (Table S2, Supplementary Material) [31], [32]. The system with best performance was selected for further evaluations. Amplitude (30 to 90%), temperature (25 to 80 °C), extractant solution (HNO₃ in the concentrations of 1 to 3 mol L⁻¹), sample mass (50 to 200 mg), and treatment time (5 to 30 min) were also evaluated. The morphology and particle size of the seaweed sample before and after UAE was evaluated by SEM and laser diffraction, respectively.

Silent conditions (without US) were carried out under the same conditions as the optimized UAE method, but either with mechanical agitation (magnetic or mechanical stirring at 500 rpm), or with no stirring, to confirm the influence of US energy on the extraction.

2.4.3. Accuracy

The evaluation of accuracy of the optimized UAE method was performed by comparison of the results with those obtained by MAWD (reference method). Furthermore, the optimized UAE method was applied to a CRM of aquatic plant (BCR60). Analyte addition experiments were also performed at three levels (50, 100 and 150% of analyte concentration present in the samples) and recovery was assessed by ICP-MS.

3. Results and discussion

3.1. Influence of ultrasound system and frequency on UAE

The UAE process can be influenced by the energy delivered by the US system [32]. For this reason, different bath systems operating at different frequencies and power (US baths A, B and C) were compared for As, Cd, Pb, Mn, Sr and V extraction from seaweed. This evaluation was performed using 50 mg of sample A and 10 mL of 1 mol L⁻¹ HNO₃ as extractant. The mixture was sonicated during 30 min at 30 °C, with the systems operating at 70% amplitude.

As shown in Fig. 1, no statistical difference was observed in extraction efficiency between bath systems operating at up to 45 kHz for all analytes (ANOVA, 95% confidence). For these systems, extraction efficiencies of up to 93% were obtained for Cd, Pb, Mn and Sr. For As and V, however, lower efficiencies were obtained, up to 75% for As and 87% for V.

Fig. 1.

Influence of US bath system and frequency on As (), Cd (), Pb (), Mn (), V () and Sr () extraction efficiency. Conditions: 50 mg of sample A, 10 mL of 1 mol L⁻¹ HNO₃ as extractant, sonication during 30 min at 30 °C and 70% of amplitude. Results of extraction efficiency are shown as agreement with the reference method (MAWD) and standard deviations are represented by the error bars (n = 3). The horizontal line indicates 100% extraction efficiency.

When bath systems operating at 80 or 130 kHz were used, extraction efficiencies below 76% were observed for Cd, Pb, Mn and Sr (Fig. 1), probably due to less energy being transferred to the medium when compared to the other evaluated frequencies (25, 35, 37 and 45 kHz), as evidenced by the lower acoustic density (Table S2, Supplementary Material). Higher acoustic density was observed in the baths operating at 25 kHz (37.2 ± 4.0 W L⁻¹), 35 kHz (33.7 ± 8.2 W L⁻¹) and 37 kHz (52.7 ± 4.8 W L⁻¹), in comparison with 45 kHz (10.8 ± 2.7 W L⁻¹), 80 kHz (6.97 ± 0.50 W L⁻¹) and 130 kHz (10.1 ± 1.3 W L⁻¹). This likely affected the intensity of the US effects (shear forces, acoustic cavitation and microjets), which could have caused the decrease in extraction efficiencies [19], [33], [34]. Furthermore, an increase in US frequency could impact the physical effects of US due to less agitation in the media and lower energy being released upon bubble implosion, which probably also affects the extraction of environmentally critical elements from seaweed [35], [36]. As there were no statistical differences between the baths operating at 25, 35, 37 and 45 kHz, the US frequency of 25 kHz was arbitrarily selected for further optimizations. Fig. 2 shows the acoustic pressure distribution in the selected US bath (25 kHz).

Fig. 2.

Acoustic pressure distribution (MPa) in the US bath (25 kHz, 100 W) at the operational depth.

As shown in Fig. 2, two positions with high acoustic pressure (higher than 40 x 10⁻³ MPa), appear above the US transducers, and where US effects are likely stronger [30]. In this sense, aiming for reproducible results, all experiments using this system were performed in these positions.

3.2. Influence of amplitude on UAE efficiency

The effects of US in the medium depend, among other factors, on the frequency, power and amplitude of the system, and different conditions can greatly affect extraction efficiency. For this reason, the amplitude of the 25 kHz US bath was evaluated (30 to 100%). For this evaluation, 50 mg of sample A, 10 mL of 1 mol L⁻¹ HNO₃ as extractant_, and 30 min of sonication at 30 °C were used. Figure S1 (Supplementary Material) shows the results of the amplitude evaluation. No significant differences (ANOVA, 95% confidence) were observed for extraction efficiency of As, Cd, Pb, Mn, Sr and V from seaweed samples for all evaluated amplitudes. However, when amplitudes higher than 70% were used, lower standard deviations (up to 5%) were observed in the results. This is possibly due to higher turbulence in the medium, resulting in better extraction homogeneity. Moreover, more agitation can provide a better interaction between the extractant and the sample, which can result in lower variation in the results [34], [37]. Aiming for better precision in the results, as well as lower energy consumption, the condition using 70% amplitude was maintained.

3.3. Influence of temperature on UAE

Increasing the temperature directly influences the solvent properties, causing a decrease in viscosity, surface tension, and increase in vapor pressure [17]. This is known to decrease the cavitation threshold, which can in turn reduce the implosion impact of cavitation bubbles [33]. However, the higher vapor concentrations in the medium observed in higher temperatures may result in the generation of a higher amount of cavitation bubbles. It is also reported that increasing the temperature favors UAE, since the diffusivity of the solvent in the medium is improved, increasing the contact between the sample and the extractant [38].

In this sense, the influence of the temperature was evaluated on As, Cd, Pb, Mn, Sr, and V extraction from seaweed. For this, the UAE was carried out in different temperatures (30 to 80 °C) using an US bath at 25 kHz, 50 mg of seaweed sample A, 10 mL of 1 mol L⁻¹ HNO₃ as extractant, and 30 min of sonication at 70% of amplitude. Fig. 3 shows the results obtained for the temperature evaluation in UAE.

Fig. 3.

Influence of temperature on the extraction of As (), Cd (), Pb (), Mn (), V () and Sr () from seaweed using US. Conditions: 50 mg of sample A, 10 mL of 1 mol L⁻¹ HNO₃ as extractant, sonication for 30 min at 70% of amplitude in an US bath (25 kHz, 100 W). Results of extraction efficiency are shown as agreement with the reference method (MAWD) and standard deviations are represented by the error bars (n = 3). The horizontal line indicates 100% extraction efficiency.

As can be observed in Fig. 3, the increase in temperature caused an increase in extraction efficiency of As and V from sample A. For As, an increase in extraction efficiency from 71% at 40 °C to 95% at 60 °C was observed. In the same way, for V, an increase from 81 to 92% in efficiency was observed when the temperature was increased to 50 °C. Moreover, this increase in extraction efficiencies for As and V at temperatures higher than 50 °C are probably associated with better interaction between the sample and the extractant. Furthermore, extraction efficiencies above 90% were observed for all analytes at temperatures higher than 60 °C, with no statistical differences (ANOVA, 95% confidence level) being observed between the results for 70 and 80 °C. For this reason, 70 °C was selected for further optimizations.

3.4. Influence of sample mass on UAE efficiency

The use of higher sample masses was evaluated to enhance LODs and LOQs, as well as sample representability. Therefore, sample masses of 50, 100, 150, and 200 mg were evaluated using 10 mL of 1 mol L⁻¹ HNO₃ as extractant and 30 min of sonication (70% amplitude and 70 °C) in the 25 kHz US bath. The effect of different sample masses on analyte extraction efficiency is shown in Fig. 4.

Fig. 4.

Influence of sample mass on As (), Cd (), Pb (), Mn (), V () and Sr () extraction from seaweed sample A. Conditions: 10 mL of 1 mol L⁻¹ HNO₃ as extractant, sonication for 30 min at 70 °C and 70% of amplitude in an US bath operating at 25 kHz. Results of extraction efficiency are shown as agreement with the reference method (MAWD) and standard deviations are represented by the error bars (n = 3). The horizontal line indicates 100% extraction efficiency.

As can be seen in Fig. 4, extraction efficiencies above 90% were observed for all analytes when sample masses up to 150 mg were used. When 200 mg of sample were used, Cd, Pb, Mn, V and Sr agreements were all higher than 92%. However, a decrease in extraction efficiency was observed for As (83%), which could be associated with a lower interaction of the sample with the extractant (as the sample to extractant ratio was reduced). Nevertheless, the condition using a sample mass of 200 mg was chosen for further evaluations in order to improve LODs and LOQs of the method.

3.5. Influence of extractant concentration on UAE efficiency

Aiming to increase the extraction efficiency for As, different HNO₃ concentrations (1, 2, and 3 mol L⁻¹) were evaluated. For this, UAE was carried out using previously optimized conditions: 200 mg of sample A, 10 mL of HNO₃ solution as extractant, and sonication for 30 min at 70 °C and 70% of amplitude in an US bath (25 kHz, 100 W). The results are shown in Fig. 5.

Fig. 5.

Influence of HNO₃ concentration on the extraction of As (), Cd (), Pb (), Mn (), V () and Sr () from seaweed. Conditions: 200 mg of sample A, 10 mL of HNO₃ as extractant, sonication for 30 min at 70 °C and 70% of amplitude in an US bath operating at 25 kHz. Results of extraction efficiency are shown as agreement with the reference method (MAWD) and standard deviations are represented by the error bars (n = 3). The horizontal line indicates 100% extraction efficiency.

As shown in Fig. 5, an increase in As extraction was observed with the increase in HNO₃ concentration (up to 95%). However, there was no statistical difference (t-test, 95% confidence) in the results for As, Cd, Pb, Mn, Sr, and V when 2 or 3 mol L⁻¹ HNO₃ were used for the UAE. Therefore, looking to keep the concentration of reagents as low as possible and to follow the green chemistry principles, 2 mol L⁻¹ HNO₃ was selected as extractant.

3.6. Influence of extraction time on UAE efficiency

Sonication time was also evaluated for the UAE procedure. For this evaluation, UAE was performed for 5 to 30 min using an ultrasonic bath (25 kHz, 100 W), 200 mg of sample A, 10 mL of 2 mol L⁻¹ HNO₃ as extractant, 70 °C, and 70% of amplitude. The results for the evaluation of extraction time are shown in Fig. 6.

Fig. 6.

Influence of treatment time on As (), Cd (), Pb (), Mn (), V () and Sr () extraction. Conditions: 200 mg of sample A, 10 mL of 2 mol L⁻¹ HNO₃ as extractant, sonication at 70 °C and 70% of amplitude in an US bath operating at 25 kHz. Results of extraction efficiency are shown as agreement with the reference method (MAWD) and standard deviations are represented by the error bars (n = 3). The horizontal line indicates 100% extraction efficiency.

As the extraction time was reduced from 30 min (Fig. 6), a reduction in extraction efficiency was observed, mainly for As, Mn, Sr and V, with efficiencies below 90%. Interestingly, treatment time did not have a significant influence on Cd extraction (efficiency higher than 95% for all conditions), and only had a significant influence on Pb extraction when the 5 min condition was used (being reduced to an efficiency of 89.8 ± 3.3%). Hence, an extraction time of 10 min would be enough to obtain quantitative results for Cd and Pb. When considering only Cd, a treatment time as low as 5 min could be used. However, as a compromise condition in which all analytes had an extraction efficiency higher than 90%, the 30 min treatment time was kept. In this sense, the best UAE conditions for As, Cd, Pb, Mn, Sr, and V extraction from seaweed samples and further determination by ICP-MS were: 2 mol L⁻¹ HNO₃ as extractant, 200 mg of seaweed sample, and sonication during 30 min at 70 °C and 70% amplitude in a 25 kHz US bath (acoustic density of 37.2 ± 4.0 W L⁻¹).

3.7. Comparison of UAE with silent conditions

In order to confirm the effect of ultrasound energy on As, Cd, Pb, Mn, Sr, and V extraction from seaweed, three different silent conditions were performed (using either magnetic or mechanical agitation at 500 rpm, and no agitation). The procedures were performed using the best UAE conditions (200 mg of sample, 10 mL of 2 mol L⁻¹ HNO₃ as extractant, 70 °C and 30 min of extraction). The comparison between the different silent conditions and the UAE method (optimized conditions) is shown in Fig. 7.

Fig. 7.

Results obtained for silent conditions: no agitation, magnetic stirring (500 rpm) or mechanical agitation (500 rpm); and optimized UAE for As (), Cd (), Pb (), Mn (), V () and Sr () extraction. Extraction conditions: 200 mg of sample A, 10 mL of 2 mol L⁻¹ HNO₃ as extractant, 30 min of extraction at 70 °C. For UAE, a US bath (25 kHz, 100 W) operating at 70% amplitude was used. Results of extraction efficiency are shown as agreement with the reference method (MAWD) and standard deviations are represented by the error bars (n = 3). The horizontal line indicates 100% extraction efficiency.

It was observed that extraction efficiencies better than 95% for all analytes were only possible when UAE was used. With the use of US, an increase in extraction efficiency of up to 20% for As, 16% for Cd, and 29% for V was observed in comparison to the silent conditions. Furthermore, high relative standard deviations (RSDs) were observed for some elements when the silent conditions were used, while RSDs for the optimized UAE method were below 5% for all analytes.

For mechanical agitation, possible contamination was observed for all analytes, which presented relatively high blank values (Table S3, Supplementary Material). Cd and Pb blank values for this condition were so high that the concentration of these elements was not detectable in seaweeds, and overestimated extraction efficiencies were observed for Mn (120 ± 15%) and V (130 ± 8%). This could be a result of contamination caused by leaching of the metallic rotating tip due to direct contact with the reaction mixture. In this way, due to the high contamination mechanical stirring is not feasible for extraction of the analytes from seaweeds.

A reduction in particle size was observed after UAE (72.8 ± 3.3 µm) in comparison to the original sample (327 ± 43 µm). Furthermore, particle morphology was also evaluated by SEM before and after UAE (Fig. 8).

Fig. 8.

SEM images of Nori seaweed sample: A) Before applying the proposed UAE method; B) After applying the proposed UAE method.

As can be observed in Fig. 8, fragmentation was observed on the seaweed surface after UAE, being likely related to microjets near the sample surface, which could result in better extraction efficiency compared to the silent conditions [36], [39].

3.8. Analytical figures of merit and application of the proposed UAE method

The accuracy of optimized UAE method was evaluated using a CRM of aquatic plant (BCR60), by analyte addition studies (at 50, 100, and 150% of the analyte concentration present in the samples), and by comparing the results with those obtained by the reference method (MAWD). Table 1 shows the certified and obtained values for BCR60 by optimized UAE procedure and determination by ICP-MS.

Table 1.

Results obtained by ICP-MS for As, Cd, Pb, Mn, Sr, and V in CRM BCR60 after the optimized UAE method (values in µg g⁻¹, mean value ± standard deviation, results in triplicate).

Analyte	UAE	Certified value
As	5.87 ± 0.48	n.i.
Cd	2.05 ± 0.51	2.20 ± 0.10
Pb	52.5 ± 6.8	63.8 ± 3.8
Mn	1573 ± 231	1759 ± 51
V	3.02 ± 0.40	n.i.
Sr	172 ± 26	n.i.

n.i.: not informed

For the CRM analysis (Table 1), no difference was observed between the results obtained by ICP-MS after UAE and the certified values (t-test, 95% confidence level). Furthermore, analyte addition studies resulted in recoveries from 91 to 108% for all analytes at all levels. As for the comparison of UAE with the reference method (MAWD) no statistical difference (t-test, 95% confidence level) was observed for the results in any of the samples (Table 2).

Table 2.

Results obtained for As, Cd, Pb, Mn, Sr, and V determination by ICP-MS in different seaweed samples after UAE or MAWD (values are expressed as µg g⁻¹, mean ± standard deviation, results in triplicate).

Analyte	Seaweed sample
	A	B	C	D	E	F
UAE method
As	20.6 ± 0.8	15.8 ± 0.4	16.6 ± 0.7	22.2 ± 0.6	55.4 ± 1.8	20.7 ± 0.7
Cd	1.24 ± 0.04	9.22 ± 0.18	11.2 ± 0.1	1.63 ± 0.26	0.652 ± 0.030	0.500 ± 0.013
Pb	0.224 ± 0.008	<0.130a	0.163 ± 0.003	0.360 ± 0.030	0.421 ± 0.015	0.240 ± 0.022
Mn	23.6 ± 1.06	36.0 ± 1.4	44.5 ± 0.6	4.17 ± 0.17	2.75 ± 0.16	5.20 ± 0.17
Sr	25.3 ± 0.9	19.3 ± 1.0	9.58 ± 1.29	714 ± 35	988 ± 40	1192 ± 112
V	2.03 ± 0.10	1.46 ± 0.065	1.19 ± 0.01	0.804 ± 0.017	0.488 ± 0.039	0.560 ± 0.015
MAWD method
As	21.7 ± 1.9	17.0 ± 0.5	17.8 ± 0.6	24.6 ± 1.2	57.3 ± 0.6	23.2 ± 0.8
Cd	1.25 ± 0.09	9.50 ± 0.30	11.8 ± 0.3	1.56 ± 0.07	0.593 ± 0.040	0.514 ± 0.031
Pb	0.225 ± 0.015	<0.158b	0.160 ± 0.008	0.322 ± 0.011	0.412 ± 0.024	0.222 ± 0.018
Mn	24.6 ± 2.4	38.9 ± 1.2	46.8 ± 0.9	4.60 ± 0.30	2.51 ± 0.26	5.69 ± 0.11
Sr	25.4 ± 2.3	20.9 ± 0.8	8.82 ± 0.10	680 ± 88	964 ± 56	1148 ± 31
V	2.13 ± 0.23	1.60 ± 0.40	1.35 ± 0.09	0.840 ± 0.050	0.465 ± 0.052	0.628 ± 0.055

^aLOQ obtained after UAE and determination by ICP-MS.

^bLOQ obtained after MAWD and determination by ICP-MS.

The LODs and LOQs of the optimized UAE and MAWD methods (in µg g⁻¹) are presented in Table S4 (Supplementary Material). It was observed that the LOQs for As, Pb, Mn, Sr and V are similar for both methods, even though the UAE method used a lower sample mass. These low LOQ values allow the determination of As, Cd, Pb, Mn, Sr and V at trace levels. Additionally, the LOD and LOQ values for As, Pb, Mn and Sr obtained for the proposed UAE method were lower than LODs and LOQs of the reference method, which allow for quantification in lower concentrations, an advantage when biomonitoring environmentally critical elements.

It is known that C concentrations in solution higher than 250 mg L⁻¹ can cause interferences in determination by ICP-MS, due to charge transfer reactions with hard to ionize elements such as As, as well as other matrix and spectral interferences [40]. The dissolved C concentration obtained for extracts after the UAE method was 4384 ± 311 mg L⁻¹, while C concentration for the MAWD method was 208 ± 39 mg L⁻¹. However, this issue can be avoided by sample dilution. In this sense, although a higher C concentration was observed for the UAE extracts in comparison to MAWD digests (which is to be expected for extraction procedures) [41], the determination by ICP-MS was performed with no interferences when a previous 20-fold dilution of the extracts was carried out in 5% HNO₃.

The optimized UAE method was also applied to different seaweed samples, including samples from the Antarctic region (Table 2). Analyte concentrations varied greatly among different seaweed species. For Antarctic macro algae samples (B to F), As concentrations ranged from 15.8 to 55.4 µg g⁻¹, Cd from 0.500 to 11.2 µg g⁻¹, Pb from < 0.130 to 0.421 µg g⁻¹, Mn from 2.25 to 44.5 µg g⁻¹, V from 0.488 to 1.46 µg g⁻¹, and Sr from 9.58 to 1192 µg g⁻¹.

The observed values are within the range of concentrations found in previous studies, which also reported varying concentrations for As (4.00 to 447 μg g⁻¹), Cd (0.013 to 12.7 μg g⁻¹), Mn (0.330 to 449 μg g⁻¹), Pb (0.104 to 21.5 μg g⁻¹), and V (0.500 to 38.4 μg g⁻¹) in Antarctic seaweed samples [4], [42], [43], [44], [45]. In the same way, results obtained for sample A (commercial edible Nori) are in agreement with a previous study on the same seaweed species [46].

4. Conclusions

The proposed UAE procedure was demonstrated to be a successful sample preparation method of seaweed samples for further As, Cd, Pb, Mn, Sr, and V determination by ICP-MS. The optimized UAE method allowed to avoid the use of concentrated acids and only a 2 mol L⁻¹ HNO₃ solution was required as extractant in a relatively short extraction time (30 min), using a 25 kHz US bath (with 7.43 ± 0.80 W of output power and 37.2 ± 4.0 W L⁻¹ of acoustic density) at 70 °C and 70% of amplitude. In comparison with the conventional MAWD method, UAE was about 2.2 times quicker. The UAE method is also simpler and safer than MAWD, as only diluted reagents were employed and overall operational conditions are milder (lower temperature and vessels are not pressurized during the procedure). Furthermore, the proposed UAE method presented higher detectability (as can be observed by the lower LOQ values for almost all analytes). In view of the advantages, the proposed UAE method was considered better than the reference MAWD method for further environmentally critical elements determination.

It should be noted that only when US was applied, quantitative extraction was observed for all analytes in the seaweed samples (extraction efficiencies better than 95%). When the silent condition using mechanical stirring at 500 rpm was applied, contamination was observed for all analytes. Furthermore, extraction efficiencies below 90% were observed for As, Cd and V using magnetic stirring (500 rpm) and for the condition without any stirring. Moreover, the obtained standard deviations in the silent conditions were higher than those obtained for UAE. Additionally, the UAE method demonstrated high accuracy by CRM (BCR60) analysis, analyte addition experiments at three levels, and comparison of results with those obtained by a reference method. Finally, the proposed UAE method can be considered as a promising alternative for environmentally critical elements biomonitoring in seaweed.

CRediT authorship contribution statement

Gustavo Gohlke: Writing – original draft, Validation, Methodology, Investigation, Formal analysis, Conceptualization. Vitoria H. Cauduro: Writing – review & editing, Methodology, Investigation, Formal analysis, Conceptualization. Emanuele Frozi: Methodology, Investigation, Formal analysis. Luana F. Rocha: Investigation, Formal analysis. Giancarlo R. Machado: Methodology, Investigation, Formal analysis. Alessandra S. Henn: Methodology, Investigation, Formal analysis, Conceptualization. Yang Tao: Methodology, Investigation. Marcia F. Mesko: Methodology, Investigation. Erico M. M. Flores: Writing – review & editing, Conceptualization, Funding acquisition, Supervision.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

The following are the Supplementary data to this article: