Introduction

The elemental analysis makes it possible to determine the contents of a given chemical element in a matrix, independently of its chemical speciation. Many analytical tools exist, depending on the nature and range of concentrations sought. In this field, the spectrometric methods have particularly developed. Increasingly sensitive techniques made it possible to accurately detect elements at trace or ultra-trace levels. The metallic trace elements are naturally occurring in soils, sediments, and rocks at levels < 0.1%. Some of them are essential to life, but all are potentially toxic from a certain dose, the toxicity also depending on their chemical speciation [1]). The massive use of these elements by humans led to an increase in their levels in the various environmental compartments (air, soil, water, sediment, biota, etc.). The health and environmental impacts related to these pollutants and their transfers are the subject of numerous studies, to get a better knowledge of the contents in different environments, or of the mechanisms that link chemical speciation, mobility, toxicity, etc. In the literature, the frequently used analytical tools are techniques such as AAS (atomic absorption spectrometry), ICP-MS (inductively coupled plasma mass spectrometry), and ICP-OES: inductively coupled plasma optical emission spectrometry also known as ICP AES inductively coupled plasma atomic emission spectrometry. Many standardized procedures for elementary analysis also refer to them; thus, reference ICP OES is widely used in the routine analysis in different fields such as geochemical, environmental, food, and fuel [2]. The interest of the ICP OES or ICP-MS techniques is related to their multi-element processing, unlike the AAS. On the other hand, analyses by ICP-MS or ICP OES are more expensive due to higher costs of the devices and operating conditions (i.e., consumption of argon for the formation and maintenance of plasma for ICP OES and ICP-MS). Since 2011, the appearance of the MP-AES (microwave plasma atomic emission spectrometry also named microwave-induced plasma optical spectrometry) has complemented this range of tools. This instrument, with an affordable acquisition cost, enables a multi-element analysis of major element and trace elements. It avoids the costs/hazards associated with the use of gases (flammable and/or expensive gases, handling of gas cylinders) since the plasma is obtained from N2, which can be produced from compressed air by a nitrogen generator connected to the device. The use of an auto sampler also makes it possible to optimize the analytical costs. The acquisition of data in ICP OES is simultaneous: it is, therefore, rare to use this tool for the analysis of only one or a few elements. Similarly, it is rare to start the apparatus for a small number of samples concerning the cost related to the formation and maintenance of the plasma. Conversely, the MIP OES operates in sequential mode and has almost no cost in terms of gas. Its use for the analysis of one or a few elements and/or few samples is more common. However, the sequential measurements can constitute one drawback as that the time and solution volume spent in processing them is separate for each metal, as noticed by Vural et al. [3].

Since 2012, several studies have investigated the performance of this new tool with various applications in terms of elements and matrices [4]. Table 1 reports the nature of matrices and elements for some studies in which at least one of the following elements Cd, Cr, Cu, Ni, Pb, and Zn was analyzed by MIP OES. The use of MIP OES is increasingly noted in the literature, for extremely diverse applications (Table 1). The technique was applied for the analysis of foodstuff, textiles, plants, fuels, soils, sediments, ores, etc. However, some of these studies do not indicate the operating conditions and do not provide any evidence as to the quality of the results obtained. Thus, detailed comparative studies in terms of procedures and performances have been carried out on various matrices, but often on a limited number of analytes. Studies comparing the results obtained by MIP OES and other techniques are, however, rather few, particularly when focusing on the elements Cd, Cr, Cu, Cd, Ni, Pb, and Zn. A few studies offer comparisons with the results obtained by ICP OES [5, 6], but again, the number of analytes and/or matrices and is often restricted.

Table 1 Nature of the samples and elements analyzed (including at least one of the following elements: Cd, Cr, Cu, Ni, Pb, and Zn) with MIP OES based on the literature study

To our knowledge, fired brick has never been analyzed by MIP OES and the analysis of sewage sludge was only reported in an application note from the manufacturer (but Cd, Ni, and Pb were not measured) [7]. In this paper, our interest focuses on the analysis of Cd, Cr, Cu, Ni, Pb, and Zn in acid digests of various matrices: marine/estuarine sediments, sewage sludge, and fired brick. Ketheesan et al. [8] mentioned the analysis of the residual sewage sludge obtained after a sequential extraction procedure prepared, but only Fe was analyzed and not the other elements analyzed in our study. We conducted the analyses in parallel using MIP OES and ICP OES, to compare the methods and performances of both techniques. Tables 1 and 2 show that the different studies were performed with various instruments with different wavelengths. However, the purpose of our investigation was to close gaps that were revealed in the literature.

Table 2 Wavelengths selected for the elements (Cd, Cr, Cu, Ni, Pb, and Zn) analyzed with MIP OES based

Materials and methods

Samples

Three different materials were analyzed to obtain the total content in Cd, Cr, Cu, Ni, Pb, and Zn. A sewage sludge was collected from a sewage treatment plant in Normandy, France. The sludge was sampled in the greenhouse and disposed after centrifugation for further and final drying. Back to the laboratory, the sludge dried at 40 °C until constant weight. The dried sludge then grinded for 15 min using an IKA crusher (IKA® Tube Mill Control) and finally stored in polyethylene vessels at 4 °C.

River sediment collected from the Seine River (France), near the harbor of Rouen. The sediment was homogenized, air-dried for 4 days, and grinded manually using an agate pestle and mortar, due to the particle solidity. The final sample materials stored at 4 °C for the further analysis of fired bricks (third material). Two samples were analyzed: BC (commercial fired brick) manufactured in France while BL (brick prepared in the laboratory). Both bricks were grinded and stored at 4 °C.

Certified reference materials were also used to ascertain the accuracy and precision of the methods. The HR-1 CRM is harbor sediment from the mouth of the Humber River (Ontario) and was obtained from Environment Canada (but Cd is non-certified for this one). The TH-2 CRM (purchased from Environmental and Climate Change Canada) is harbor sediment collected in Lake Ontario. The CRM031 is domestic sewage sludge purchased from Sigma-Aldrich.

Samples preparation

All the materials were digested before the analysis, microwave-assisted acid digestions performed in Teflon® reactors using a Berghof Speedwave MWS-2 microwave oven, and all the reagents were in analytical grade. The deionized water with a resistivity of 18.2 MΩ was produced by a Milli-Q water system (MAXIMA). A standard stock solution of 1000 mg L−1 (Agilent) was used for all the elements calibration. Glassware and plastic materials (including the 50 mL polyethylene vessels used for storage) were pre-soaked for 24 h in 10% v/v HNO3 and rinsed with deionized water.

Digestion procedure applied to the samples and certified materials: 0.2 g (sediment) or 0.5 g (sludge, brick) of dried sample was digested with 10 mL of HNO3 (sludge) or of aqua regia (HCl/HNO3 v/v = 3:1) (sediment and brick). The heating program consisted of 3 (sludge) or 4 (sediment, brick) steps with maximum temperature up to 175 °C for 10 to 15 min. The obtained digests were then transferred and completed to 50 mL in a volumetric flask with deionized water. After filtration at 0.45 μm, the digest solutions were stored in polyethylene vessels at 4 °C before chemical analysis. Each digestion was performed in triplicate, and blank prepared with the same medium. Each replicate of each acid digest obtained from the various samples (sediment or brick or sludge or CRM) was furtherly analyzed by MIP OES and by ICP OES.

Instrumental apparatus

Determinations of the Cd, Cr, Cu, Ni, Pb, and Zn in the digests previously prepared were performed using a microwave-plasma atomic emission spectrometer (MIP OES) and an inductively coupled plasma atomic emission spectrometer (ICP OES). An Agilent 5100 ICP OES was used for the determination of the elemental concentrations. The operating conditions are reported in Table 3.

Table 3 Operating conditions of the analytical instruments

The used microwave-plasma atomic emission spectrometer was an Agilent 4200 MIP OES. The N2-based plasma was operated from the compressed air supply and a nitrogen generator (Agilent 4107). An autosampler (Agilent SP4), an OneNeb concentric nebulizer, and a double pass spray chamber were used. The instrument software (MP Expert) was allowed to optimize automatically some instrumental parameters (i.e., torch alignment and wavelength calibration by using a wavelength calibration solution (Agilent) as well as nebulizer gas pressure and was also used for automatic background correction or FLIC (Fast Linear Interference Correction) correction, when it was necessary. The MIP OES conditions are summarized in Table 3.

For both devices, each elemental concentration was estimated from its respective calibration curve. A multi-elemental standard solution (Agilent) was used to prepare the calibration solutions (concentration range 0–5 or 10 mg L−1 depending on elements and instruments, in 5% v/v HNO3).

Results and discussion

In Table 3, the wavelengths are selected for each element and technique, to get the best sensitivity for the minimum interferences. The same wavelengths were selected for MIP OES and ICP OES (except for Cr and Pb, optimized routinely with different sensitivity and possible interferences); the sensitivity of the elements depends on many instrument parameters and matrices. The wavelengths selected for MIP OES are the most frequently reported in the literature (Table 2). Some other lines are sometimes chosen, i.e., 226.502 nm for Cd and 368.346 nm for Pb [13], 405.781 nm [2], or 481.053 nm [14] for Zn. Two lines per element can be selected for the measurements, particularly when interferences were suspected. Similarly, Karlsson et al. [9] selected two lines for each element they analyzed. They discarded the Pb line at 283.305 nm that presented interferences with Cr and Fe, which then required a FLIC correction. The analytical lines were chosen for optimal sensitivity using certified reference materials matrices (sediment, sludge, and brick).

The calibration curve range was 0–5 mg L−1 for each element analyzed with MIP OES with ICP OES. A linear or rational calibration curve could be selected for each instrument, according to the R2 values. The linear curves were chosen without any exception for both instruments. The R2 values obtained for the linear calibration range are reported in Table 4 and were greater than 0.9984 for all analyses, as described also by Niedzielsky et al. [10].

Table 4 Limits of detection and quantification calculated by calibration curves (R2, LOD, and LOQ in mg L−1)

Limits of detection and limits of quantification

For the ICP OES and MIP OES, the instrumental limits of detection (LOD) and limits of quantification (LOQ), for the different elements of interest, were estimated via the calibration curves approach, as well as were obtained using certified standard materials with the same medium. The LOD and LOQ were calculated as 3*σ/a and 10σ/a of calibration graph, respectively, with σ being the standard deviation of the regression line/intercept and a is the slope of the calibration curve [13, 60]. The LODs and LOQs values are reported in Table 4.

As reported by Li et al. [5], the LOD and LOQ values estimated by both techniques were comparable, suggesting comparable performances for both devices. Moreover, it can be noticed that, except for Cd, the estimated LOD and LOQ for MIP OES are lower (or very close to for Cu) than for ICP OES. Such observations, for various elements, have been reported by several authors: Li et al. [5] for Mn, Zhao et al. [6] for Ni and Pb, or Donati et al. [2] for Cr, Ni, and Pb. Donati et al. [2] proposed several hypotheses to support such observations. They mentioned the influence of the integration time and sample introduction/nebulization system. They also referred to the fact that contrary to ICP OES, that collects emission signals simultaneously, with a radial and axial view, then at one plasma position for all the lines and elements, the MIP OES allows collecting emission signals at different (then optimized) position in the plasma for each emission line (atomic and ionic lines realized to obtain the concentrations). As already reported by previous authors, the MIP OES is not superior performance than ICP OES. The nitrogen plasma temperature is lower (5000 K) compared with argon plasma (6000–10,000 K), and the resulting lower sensitivities. Also, the MIP OES detection system is more straightforward than ICP OES. The commercial MIP OES instrument is based on a Czerny–Turner monochromator (600 mm focal length) and a back-thinned Peltier-cooled charge-coupled device (CCD) detector. The MIP OES instrument has been optimized for many analyses in large material matrices and routine applications.

Samples analysis

Analysis of certified reference materials

The analysis of the three certified reference materials was performed to test the accuracy of the methods. The results determined were satisfying for both instruments. The detected values were generally within the uncertainty limits of the certified values at 95% confidence level (Fig. 1). Mainly, such results validate the various digestion procedures applied; they also confirm the analytical techniques. Indeed, even if all measured values were of the same order of magnitude than the certified values, it could be observed that MIP OES slightly overestimated Cr levels in the sludge and TH2 CRM, overestimated Pb. Meanwhile, it underestimated Zn in the sludge, whereas ICP OES overestimated Cu and Zn in the sludge. Such differences could be attributed to the analytical instrument, related to the ionization mode, parameters, etc. However, the recoveries are satisfying for all the tested elements and CRM. Finally, the results demonstrated the ability of the MIP OES to accurately determine Cd, Cr, Cu, Ni, Pb, and Zn in estuarine or marine sediments and sewage sludges; the t test (p < 0.05) was performed for all the simple analysis.

Fig. 1
figure 1

Tests using certified reference materials a sludge; b TH2 sediment; c HR1 sediment (the found values by ICP OES or MIP OES are the mean ± SD of three parallel determinations; variations between instruments were statistically significant (p < 0.05))

Analysis of samples of sediments, bricks, and sewage sludge

The various samples (domestic sludge, estuarine sediment, and fired brick) were analyzed, after acid digestion, by both MIP OES and ICP OES. The results are reported in Table 5.

Table 5 Comparison of analyses for different samples by MIP OES and ICP OES

The data obtained by both instruments and for all the tested samples (CRM included) were statistically analyzed. One-way ANOVA statistical analysis was performed to evaluate the statistical difference between the two sets of data obtained by MIP OES and ICP OES. Table 5 reports the probability of p values. The acquired results from MIP OES and ICP OES are considered different with p < 0.05.

The obtained results from MIP OES were very close to those obtained by ICP OES, which suggested that the analysis performance of MIP OES and ICP OES was comparable for the elements analyzed and the various matrix tested, as it was already reported. Donati et al. [2] concluded that the performance of MIP OES for determining Cr, Ni, Pb (and V) in gasoline was comparable if not better than ICP OES. Li et al. [5] also concluded to comparable performance of the two techniques, with respect to the tested matrices (animal feed and fertilizers digested or extracted in acids) for Zn and Cu, even if some individual p values were found < 0.05 (for one sample for Cu and one sample for Zn). Zhao et al. [6] compared the performance of MIP OES and ICP OES for the analysis of fur and leather acid digests, for the same trace elements Cd, Cr, Cu, Ni, and Pb (Zn excluded). They concluded to comparable performances even if, again, they noticed individual statistically significant differences for the two techniques for Pb (in two samples) Cu (in one sample) and Cr (in four samples), that they attributed to the inhomogeneity of their samples. Jung et al. [60], also compared the performance of the two instruments, but only for one element (Mn in wine) and also concluded that the sensitivity of MIP OES and ICP OES was comparable. We demonstrated that MIP OES elemental analyses are realizable in different matrices such as sludge, sediment, and bricks, with high efficacy. Although the plasma is less robust than an ICP OES, which requires matrix-matching calibration methods (e.g., standard additions) to ensure adequate accuracy when analyzing complex matrix samples [61, 62], MIP OES may be considered an attractive alternative to ICP OES for tailored applications when both proper operational conditions and calibrations strategies are adopted [63].

Conclusion

This study compared the concentrations of Cd, Cu, Cr, Ni, Pb, and Zn which were performed by ICP OES and MIP OES, in which acids (HNO3 or aqua regia) digested the samples of various nature (bricks, sediments, and sewage sludges). Similar performances of both devices were established for the studied trace elements, with better-observed LOD for MIP OES regarding all the tested elements, except for Cd. The results obtained by MIP OES and ICP OES were very comparable, and the differences somehow observed were acceptable. MIP OES appeared as ultimately suitable to analyze the trace elements (Cd, Cu, Cr, Ni, Pb, and Zn) in acid digests of sewage sludge and fired bricks. The cost reduction that is partly related to the absence of argon consumption is an added advantage, confirmed MIP OES as a good alternative instrument to ICP OES.