Quantification of Linear Alkylbenzene Sulphonates in Complex Sludge Samples: Influence of Matrix Effects in Calibration Methods

Microchemical Journal 204 (2024) 111089 Available online 28 June 2024 0026-265X/© 2024 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/bync-nd/4.0/). Quantification of linear alkylbenzene sulphonates in complex sludge samples: Influence of matrix effects in calibration methods Julia Martín a , * , Carmen Mejías a , Juan Luis Santos a , Irene Aparicio a , Esteban Alonso a , John Heinze b a Departamento de Química Analítica, Escuela Polit´ ecnica Superior, Universidad de Sevilla. E-41011 Seville, Spain b Council for LAB/LAS Environmental Research (CLER), 529 14 th Street, NW, Suite 1280, Washington, DC 20045, USA ARTICLE INFO Keywords: Linear alkylbenzene sulphonate Calibration methods Matrix effect Sludge samples ABSTRACT Linear Alkylbenzene Sulfonates (LAS) are among the priority organic pollutants in sludge owing to their high concentrations, due to its use as a major cleaning agent (surfactant) in laundry detergents and household cleaning products, and to the concentration limits of 2,600 mg/kg outlined in the EU directive draft for land application of sludge. In this work, a method has been optimized and validated for the accurate quantification of LAS homologues C10-C13 in different types of samples in the sludge treatment and disposal process including primary, secondary, and digested sludge, compost, and soil. The type of matrix had a significant impact on the method’s validation process, resulting in the need of the use of distinct dilution factors for each matrix. Overall, the procedure propose involves a cost-effective and user-friendly technique that combines both extraction and clean-up in a single step (ultrasound-assisted extraction and dispersive solid phase extraction). Analytical determination is conducted using liquid chromatography-tandem mass spectrometry. The method demonstrated excellent accuracy (relative recoveries exceeding 84.5 % across all types of sludge samples) and the limit of quantification enables the measurement of LAS concentrations up to 1000 times lower than the concentration limits stipulated in the EU directive draft. For an accurate quantification, three calibration approaches were tested (external calibration, standard addition, and matrix-matched calibration). Standard addition was the preferred method for quantitative analysis. It helps mitigate matrix effects that could otherwise distort the accuracy of analyte measurements, making it crucial for quality control, especially in monitoring applications where matrix effects may introduce bias into concentration calculations. 1. Introduction Within the European Union (EU), the typical and prevalent practice for sewage sludge disposal involves its utilization as an organic soil amendment. This choice is not only cost-effective but also environmentally beneficial, as it provides a valuable source of organic matter, nitrogen, and phosphorous, making it a useful fertilizer and soil conditioner [1]. However, this practice comes with potential risks due to the presence of heavy metals and organic pollutants at high concentration levels in sludge. These contaminants tend to accumulate in sludge and, subsequently, can find their way into the soil [2]. For instance, Gonzalez et al. (2012) [3] found concentrations of Linear Alkylbenzene Sulfonates (LAS) in compost-amended soil up to 18.7 mg/kg dry weight (dw). LAS are a major cleaning agent (surfactant) used in laundry detergents and household cleaning products. The EU Directive 86/278/EEC [4] has regulated the use in agriculture of residual sludge from domestic and urban wastewater but does not regulate the concentration of organic pollutants. In 2000, to address the current challenges in sludge management and mitigate the presence of pollutants in soils, the EU introduced the third draft of a future sludge directive titled “Working Document on Sludge” [1]. This draft marked a significant step as it introduced concentration limits for seven groups of organic compounds for the first time. Among these organic pollutants, LAS are noteworthy because their concentrations in anaerobically stabilized or anaerobic digested sludges exceeding the EU directive draft limits (2,600 mg/kg) [5]. Currently, regulations regarding LAS limits in sludge are not standardized across EU member states, Denmark has implemented rigorous measures by establishing a LAS limit of 1,300 mg/kg dw, whereas Belgium has set its limit slightly higher at 1,500 mg/ kg dw [6]. Furthermore, the scientific community has mainly focused * Corresponding author at: Departamento de Química Analítica, Escuela Polit´ ecnica Superior, Universidad de Sevilla, c/ Virgen de ´ Africa, 7, 41011 Seville, Spain. E-mail address: [email protected] (J. Martín). Contents lists available at ScienceDirect Microchemical Journal journal homepage: www.elsevier.com/locate/microc https://doi.org/10.1016/j.microc.2024.111089 Received 7 May 2024; Received in revised form 18 June 2024; Accepted 27 June 2024 Microchemical Journal 204 (2024) 111089 2 study on emerging pollutants [7]. Hence, the imperative lies in reliable and easy-to-perform analytical methods for the precise quantification of different homologues of such pollutants in complex sludge samples. According to World Health Organization, the percentage distribution of LAS homologues commonly used in laundry detergents in Europe is as follows: C10 (10–15 %), C11 (25–35 %), C12 (25–35 %), C13 (15–30 %) [8]. Analytical methodologies developed earlier predominantly concentrated on the multiresidue determination of different pollutants, resulting in minimal endeavors to validate simple, automated, and environmentally friendly methods for Table 1 Optimized parameters of the MS/MS determination after LC separation. Compound Precursor ion (m/z) MRM 1 a MRM 2 b Fragmentor (V) Collision energy (eV) LAS C10 297 [M−H] - 183 119 180 36/50 LAS C11 311 [M−H] - 183 119 175 36/50 LAS C12 325 [M−H] - 183 119 180 40/50 LAS C13 339 [M−H] - 183 119 180 40/50 a : Quantitation ion; b : Confirmation ion. Fig. 1. Influence of clean-up sorbent on the LAS recovery (n =3). Fig. 2. Influence of extraction technique on the LAS recovery (n =3). Fig. 3. Global response surface plots, corresponding to the desirability function, when optimizing the following pair of factors from the extraction procedure: (cycles vs. sorbent amount (A), sample amount vs. sorbent amount (B), and cycles vs. sample amount (C)). Results were evaluated using a 95 % confidence interval. J. Martín et al. Microchemical Journal 204 (2024) 111089 3 determining LAS. For liquid chromatography (LC) determination, less expensive ultraviolet (UV) detection has been an option [3]. However, the higher amount of noise in the LC-UV system affects selectivity, accuracy and sensitivity and is the reason why it has been widely replaced by mass spectrometry (MS) analyzers such as tandem mass spectrometry (MS/MS). However, electrospray ionization can introduce matrix effects [9,10], particularly when co-eluting interferences are present in the matrix extract along with target analytes. It is of utmost importance to consider these matrix effects [11], as they can potentially alter analytical results. Several strategies exist to mitigate matrix effects [12,13], such as sample extract dilution [14], albeit at the cost of increased detection limits, sample extract clean-up, or the use of internal standards such as 13 C 2 -LAS-C12, which can be costly and challenging to acquire. Another approach to address matrix effects involves the utilization of various calibration methods, including matrix-matched calibration curves or the standard addition calibration method [15–17]. The objective of this research is to optimize a straightforward analytical method for the accurate quantification of LAS homologues C10-C13 commonly used in laundry detergents in Europe in complex sludge samples. The sample treatment involves a cost-effective and userfriendly technique that combines both extraction and clean-up in a single step (ultrasound-assisted extraction and dispersive solid phase extraction). Analytical determination is conducted using LC-MS/MS. The validation was performed using three different calibration strategies to determine the most suitable approach. 2. Experimental 2.1. Chemicals and reagents LC-grade acetonitrile (ACN), methanol (MeOH) and water were supplied by Romil (Barcelona, Spain). Analytical-grade ammonium acetate (≥98 %) was provided by Panreac (Barcelona, Spain). Primarysecondary amine (PSA), C18 and formic acid were provided by Scharlab (Barcelona, Spain). Florisil® was supplied by Sigma-Aldrich (St. Louis, MO, USA). Commercial LAS consist of a mixture of four alkyl chain homologues, C10 to C13, with an average alkyl chain length of approximately C11.6 [5]. The commercial sample, which contained LAS-C10 (10.3 %), LAS- C11 (35.0 %), LAS-C12 (35.1 %) and LAS-C13 (19.6 %) was supplied by CEPSA Química (Madrid, Spain). Working solutions were prepared by dilution of a stock standard solution (4000 mg/L, LAS mixture) in MeOH:H 2 O (1:1, v/v). 2.2. Instrumentation and software Chromatographic analyses were performed on an Agilent 1200 series HPLC (Agilent, USA) equipped with a vacuum degasser, a binary pump, an autosampler, and a thermostated column compartment. Three types of chromatographic columns were tested, a Zorbax Eclipse XDB-C18 (50 mm x 4.6 mm i.d., 1.8 µm particle size) acquired from Agilent (Barcelona, Spain), an Inertsil Ph-3 (150 mm x 4 mm i.d., 5.0 µm particle size) from GL Science (Dillenburgstraat, The Netherlands), and, a Kinetex Polar C18 (50 mm x 3.0 mm i.d., 2.6 µm particle size) supplied by Phenomenex (California, USA). The HPLC system was coupled to a 6410 triple quadrupole mass spectrometer equipped with an electrospray ionization source (ESI) (Agilent, USA). Ionization of analytes was carried out using the following settings: MS capillary voltage 4000 V, drying-gas flow rate 9 L/min, drying-gas temperature 350 ◦C, and nebulizer pressure 40 psi. A vortex-mixer (Velp Scientifica, Usmate, Italy), an ultrasound Sonorex digitec bath (Selecta, Berlín, Germany), a Macrotronic BLT centrifuge (Selecta, Abrera, Spain), a Dionex ASE 350 accelerated Solvent Extractor (Dionex, Sunnyvale, USA) and a sample concentrator (Stuart, Staffordshire, UK) were used for sample treatment. Statgraphics Centurion Version 18 (Statpoint Technologies Inc., Warrenton, VA, Fig. 4. External calibration, standard addition and matrix-matched calibration curves obtained for compost sample. J. Martín et al. Microchemical Journal 204 (2024) 111089 4 USA) was used for statistical treatment of data. 2.3. Sample collection and pretreatment Primary, secondary, and anaerobically digested and dehydrated sludge samples were collected from one urban WWTPs in Palomares del Río (Seville, Spain). Compost samples were collected in a 1-year sampling period collected from a composting plant where anaerobically digested sludge was treated in thermally controlled dynamic batteries with aeration provided by turning. Soil sample was Mediterranean type and was collected from an agricultural land in Seville. Samples were freeze-dried in a Cryodos-50 lyophilizer (Telstar, Spain), homogenized in a mortar, and sieved (particle size <1 mm). 2.4. Sample treatment Final optimized sample treatment was as follows: lyophilized samples (0.25 g dw) were transferred to glass centrifuge tubes with 0.5 g of florisil and were homogenized. Tubes were briefly vortex-mixed using 8 mL of MeOH and sonicated in an ultrasonic bath for 5 min. After extraction, tubes were centrifuged for 10 min at 2900 ×g. The liquid phase was transferred to a clean tube and the extraction was repeated. The combined liquid phase of two extractions was diluted with MeOH: H 2 O (1:1, v:v) to a final volume of 25 mL for compost and soil, of 100 mL for secondary sludge and of 100 mL for primary and digested sludge of which 0.5 mL were transferred to 10 mL using a 1:1, v:v MeOH:H 2 O solution. The liquid phase was filtered through a 0.22 µm cellulose syringe filter and collected in a vial for LC–MS/MS determination. 2.5. LC-MS/MS determination Chromatographic separation was carried out in an Inertsil Ph-3 (150 mm x 4 mm i.d., 5.0 µm particle size) from GL Science (Dillenburgstraat, The Netherlands) protected by an Inertsil Ph-3 Cartridge Guard Column E (10 mm x 4 mm i.d., 5.0 µm particle size) (GL Science, The Netherlands). Mobile phase was composed of 10 mM ammonium acetate (solvent B) and MeOH (solvent A). The gradient program was as follows: 0–5 min, linear gradient from 70 to 80 % of solvent A and held for 2 min, back to initial conditions in 1 min and held for 2 min for reequilibration. Flow rate was fixed at 0.6 mL/min. The injection volume was 10 µL and the column temperature was maintained at 35 ◦C. LAS homologues were analyzed in multiple reaction monitoring (MRM) mode. Two MRM transitions were selected for each LAS homologue (C10-C13) for quantification and confirmation. LAS homologues were all ionized in negative mode. The quantitation/confirmation ions and the optimized fragmentor and collision energy for each analyte are shown in Table 1. 3. Results and discussion 3.1. Method optimization The optimization of the methodology applied to the analysis of LAS in sludge samples was carried out over two steps of the sample protocol: Table 2 Method detection limits (MDLs) and method quantification limits (MQLs) of LAS C10-C13 in soil, compost and sludge samples. MDL (mg/kg dw) MQL (mg/kg dw) C10 C11 C12 C13 C10 C11 C12 C13 Soil 0.006 0.018 0.021 0.012 0.02 0.06 0.07 0.04 Compost 0.006 0.021 0.021 0.012 0.02 0.07 0.07 0.04 Digested sludge 0.246 0.846 0.861 0.453 0.82 2.82 2.87 1.51 Secondary sludge 0.015 0.042 0.036 0.018 0.05 0.14 0.12 0.06 Primary sludge 0.258 0.867 0.858 0.465 0.86 2.89 2.86 1.55 Table 3 Recovery assay, precision and trueness of LAS C10-C13 in soil, compost and sludge matrices. Soil Low level Medium level High level Spiked Found RSD Rec Spiked Found RSD Rec Spiked Found RSD Rec (mg/kg dw) (mg/kg dw) (%) (%) (mg/kg dw) (mg/kg dw) (%) (%) (mg/kg dw) (mg/kg dw) (%) (%) LAS C10 0.52 0.48 10.9 93.2 14.4 13.4 3.10 92.8 82.4 85.1 2.57 103 LAS C11 1.75 1.95 25.3 111 49.0 46.9 3.99 95.7 280 302 2.72 108 LAS C12 1.76 1.85 23.8 105 49.1 43.9 3.86 89.3 281 288 3.09 103 LAS C13 0.98 0.96 21.2 97.6 27.4 24.7 6.29 90.0 157 158 2.57 101 Compost LAS C10 4.12 4.57 2.70 111 20.6 19.7 4.42 95.5 82.4 82.4 5.55 100 LAS C11 14.0 13.9 19.7 99.5 70.0 65.8 3.67 94.0 280 284 5.77 101 LAS C12 14.0 13.9 13.0 98.7 70.2 64.6 4.18 92.0 281 290 2.87 103 LAS C13 7.84 7.57 11.6 96.5 39.2 38.2 3.91 97.5 157 167 3.83 106 Digested sludge LAS C10 824 824 1.39 100 8240 7762 5.05 94.2 16,480 16,974 3.55 103 LAS C11 2800 2778 2.65 99.2 28,000 27,076 1.98 96.7 56,000 57,120 2.95 102 LAS C12 2808 2745 0.54 97.8 28,080 26,957 3.28 96.0 56,160 57,283 2.86 102 LAS C13 1568 1631 5.18 104 15,680 17,875 2.07 114 31,360 36,064 2.81 115 Secondary sludge LAS C10 8.24 6.98 4.94 84.8 41.2 40.9 4.95 99.3 165 161 6.72 97.4 LAS C11 28.0 27.7 0.15 99.0 140 133 1.09 95.0 560 542 3.52 96.7 LAS C12 28.1 32.5 1.02 116 140 137 1.00 97.8 562 532 3.14 94.7 LAS C13 15.7 21.8 0.45 139 78.4 80.8 2.45 103 314 287 3.18 91.6 Primary sludge LAS C10 824 790 1.39 95.8 8240 7482 5.05 90.8 16,480 16,480 3.55 100 LAS C11 2800 2713 2.65 96.9 28,000 25,816 1.98 92.2 56,000 56,000 2.95 100 LAS C12 2808 2757 0.54 98.2 28,080 25,665 3.28 91.4 56,160 56,160 2.86 100 LAS C13 1568 1582 5.18 101 15,680 14,394 2.07 91.8 31,360 31,360 2.81 100 J. Martín et al. Microchemical Journal 204 (2024) 111089 5 1) the chromatographic separation and LAS determination and 2) the sample treatment (extraction and purification of analytes from the sludge matrix). 3.1.1. Optimization of the chromatographic determination Three primary variables were considered to achieve a good resolution of the peaks, adequate selectivity, and sensitivity: the analytical column, the composition of the mobile phase and the detection conditions. The optimization of the MS/MS conditions was performed by direct injections (without column) of a mixture standard solution at 1.5 mg/L in MeOH:H 2 O at a flow rate of 0.6 mL/min by using MassHunter optimizer software. After obtaining m/z data for precursor ion and product ions, different mobile phases were tested in the attempt to obtain the best ionization of compounds and to achieve higher responses. MeOH and ACN were evaluated as organic mobile phases. As aqueous solvent, deionized water with different additives (formic acid and ammonium acetate) was checked. Since LAS showed higher responses in negative ionization mode a gradient elution with MeOH (solvent A) and a buffer solution ammonium acetate (10 mM) (solvent B) was selected as mobile phase. When MS/MS conditions and mobile phase were selected, different analytical columns (Zorbax Eclipse XDB-C18 (50 mm x 4.6 mm i.d., 1.8 µm particle size), Inertsil Ph-3 (150 mm x 4 mm i.d., 5.0 µm particle size) and Kinetex Polar C18 (50 mm x 3.0 mm i.d., 2.6 µm particle size) were used to obtain good resolution of the peaks. The optimal chromatographic conditions were selected according to the peak resolution and asymmetry factors. From Fig. S1 we can see that the Inertsil Ph-3 column provided a better peak shape in a short time. Although wide-bore columns and, consequently, higher flow rates result in lower ESI efficiency and reduced MS sensitivity, the Inertsil-Ph3 widebore column (150 mm x 4 mm i.d., 5.0 µm particle size) provided better peak symmetry in a comparable run time than the Kinetex Polar C18 (50 mm x 3.0 mm i.d., 2.6 µm particle size) and the Zorbax Eclipse XDB-C18 (50 mm x 4.6 mm i.d., 1.8 µm particle size). During the optimization step, the middle-bore column presented slightly tailing peaks for LAS homologues. Furthermore, the Inertsil-Ph3 generated a higher signal for LAS homologues than the other two columns, thus making the wide-bore column the preferred choice. In addition, this column generated lower pressures (up to 200 bar) than other columns. 3.1.2. Optimization of the extraction and purification conditions First of all, purification conditions were optimized. Three sorbents of different nature, i.e. one reversed-phase sorbent (C18), one mixed mode sorbent (PSA) and a normal-phase sorbent (Florisil), were evaluated as clean-up sorbents to decrease matrix effects. The experiment was carried out in triplicate using aliquots of 1 g of compost spiked with a LAS mixture at a concentration level of 15 mg/kg dw. For optimization, a general and non-optimized sample extraction method was used at first: Aliquots of samples (1 g) were weighed into 12 mL glass vials. The samples were vortexed for homogenization in 8 mL of MeOH and centrifuged for 10 min at 4050 ×g. The solvent extract was transferred to a clean tube and subjected to a clean-up step by dSPE adding 1 g of each sorbent in each case. The tubes were shaken vigorously for 1 min and centrifuged at 4050 ×g for 15 min. The solvent was filtered through a 0.22 µm nylon and diluted up to 10 mL using water. A 10 µL aliquot of the extract was injected into the LC instrument. Fig. 1 summarizes the results obtained. LAS C10-C13 exhibited a similar behaviour, poor recoveries with PSA probably explained by their retention in the sorbent through electrostatic interactions, whereas a significant decrease of the signal was observed with C18, which can be attributed to its higher retention by partitioning as consequence of its similar interactions. Acceptable recoveries were obtained using florisil, finally chosen as the clean-up sorbent material. Florisil can remove the major polar interferences eluting at the same time as the LAS homologues. Therefore, florisil was selected as clean-up sorbent for the next optimization experiments. Different extraction techniques (ultrasound-assisted extraction (UAE followed by a clean-up by dSPE (UAE +dSPE)); UAE and dSPE combined in a single step (UAE-dSPE); and selective pressurized liquid extraction (SPLE)) were assessed using florisil as clean-up sorbent in all cases (Fig. 2). The simultaneous extraction and clean-up strategy used in UAE-dSPE and SPLE is in concordance with the fourth principle of Green Table 4 Summary of analytical methods for the determination of LAS homologues in sludge. Extraction method Sludge amount (g) Extraction solvent Solvent volume (mL) Extraction time Recovery (%) MQL (mg/kg dw) Calibration Determination Reference Soxhlet No data MeOH No data 8 h 85–99 No data EC LC-UV [22] Pressurized-liquid extraction 10 MeOH:Water (9:1, v/v) 22 3 ×5 min 52–85 0.025 EC LC-MS/MS [23] Microwaveassisted extraction 0.5 MeOH 5 10 min 83–102 0.85–4.40 (DAD) 1.11–6.09 (FLD) EC LC-UV-DAD/ FLD [24] Ultrasoundassisted extraction No data MeOH 20 3 ×20 min 78–93 0.0002–100 (MDL) EC LC-MS [25] Ultrasoundassisted extraction 1 MeOH 50 25 min 48–99 13–56 (UVDAD) 5–18 (FLD) EC LC-UV-DAD/ FLD [26] Probe sonication extraction 0.5 MeOH 10 7 min 84–97 2.78–15.22 EC LC-UV-DAD/ FLD [27] Microwaveassisted extraction 5 MeOH 25 25 min No data 0.011–0.018 EC LC-FLD [28] QuEChERS method 0.5 ACN 10 No data 89–114 0.002–1 MM LC-MS/MS [20] Ultrasoundassisted extraction 0.25 MeOH 16 2 x 5 min 85–115 0.02–2.89 SA LC-MS/MS Proposed methodology ACN: acetonitrile; EC: external calibration; DAD: diode array detector; DAD-UV: diode array detector-ultraviolet; FLD: fluorescence detector; LC-FLD: liquid chromatography-fluorescence detection; LC-MS: liquid chromatography-mass spectrometry; LC-MS/MS: liquid chromatography-tandem mass spectrometry; LC-UV- DAD/FLD: liquid chromatography-ultraviolet-diode array detector-fluorescence detection; MDL: method detection limits; MeOH: methanol; MQL: method quantification limits; MM: matrix-matched calibration; SA: standard addition. J. Martín et al. Microchemical Journal 204 (2024) 111089 6 Analytical Chemistry, integration of analytical processes and operations saves energy and reduces the use of reagents [18]. To simplify the extraction procedure, and according to the results of Fig. 2 the combination of UAE-dSPE was selected in a single step. Slightly higher extraction recoveries were achieved for the LAS mixture in addition to the significant reduction in time of the clean step in the sample protocol (mean recoveries of 81, 76 and 75 % for UAE-dSPE, UAE +d-SPE, and SPLE, respectively). The extraction time was set to 5 min in all cases to obtain homogenization of the extracted samples. Once UAE-dSPE technique and florisil sorbent were fixed as extraction technique and clean-up sorbent and, to increase extraction efficiency and greenness of the method, the sample amount, florisil amount and extraction cycles were optimized by using a Box–Behnken experimental design. The design matrix for the Box-Behnken study was generated using three levels for each of the three factors: sample amount (0.25, 0.625 and 1 g), extraction steps (1, 2 and 3) and florisil amount (500, 1250 and 2000 mg). The design matrix is shown in Table S1. A total of 15 analytical assays were carried out. All experiments were performed in a randomized order to minimize the effects of uncontrolled variables that may introduce a bias in the measurements. Three central point experiments were incorporated to estimate the experimental error that does not depend on the fitted model. To better understand the interactions of independent variables on the extraction results, the three-dimensional response surface images and corresponding contour plots were visualized. Responses for each compound in the experiments of the Box–Behnken design were first normalized between 0 and 1, and the global desirability function was defined as their geometric mean. The plot of this function: cycles vs sorbent amount (A), sample amount vs sorbent amount (B), and cycles vs sample amount (C) are shown in Fig. 3. The cycles and the sample amount resulted in the most significant variables, being noteworthy the positive interaction in the former and the negative interaction for the sample amount. When a larger quantity of sludge was used, a higher decrease in response was observed due to the considerable matrix effect. Regarding the cycles, two steps were significant to increase the extraction recoveries. According to the results, 0.25 g of sample, 0.5 g of florisil and two extraction steps were selected. The results of the final optimized method can be seen above in section 2.4. 3.2. Analytical performance 3.2.1. Matrix effect assessment Although the evaluation of the influence of different matrices of an analytical method is a mandatory step in the method’s validation, its applicability to the monitoring of trace compounds in complex sludge samples is not simple, nor is it part of the routine of most laboratories. This section focusses on the accurate procedure for quantification concentrations of LAS in different types of matrices, namely primary sludge, secondary sludge, digested sludge, compost, and soil. To assess the influence of the matrix in the quantification practice, three calibration curves were prepared for each compound: one in pure solvent (external calibration, EC), the second one using matrix-matched with commercial fertilizer samples (matrix-matched calibration, MM), and the last one using standard addition (SA) method for each type of matrix (soil, compost, and sludges). Calibration curves were constructed using the LAS peak area ratio versus the LAS homologue concentration. Calibration standards were prepared at eight different analyte concentration levels. Taking into account the LAS specific responses and the concentration levels present in sludge samples, different concentration ranges were used for each matrix (from 0.1 to 100 mg/kg dw for soil; from 5 to 800 mg/kg dw for compost and commercial fertilizer; from 10 to 1600 mg/kg dw for secondary sludge and from 200 to 240000 mg/kg dw for primary and digested sludge for the commercial LAS sample). Note that a different dilution factor was also applied depending on the type of matrix (1:200 for soil, commercial fertilizer, and compost, 1:400 for secondary sludge and 1:8000 for digested and primary sludge). For MM and SA calibrations, spiked samples were vortexed for 2 min and then left to stand for 24 h at 4 ◦C in the dark before analysis. This allows the analytes to come into full contact with the sample. Ideally, a matrix-matched standard should not contain any of the selected compounds, but in practice this is not always possible. The areas of compounds that are present in the blank extract must be subtracted to the area obtained from the matrix-matched standards. Table S2 shows the linearity assessment obtained in the working range of LAS in different types of calibrations and sludge samples. Fig. 4 shows calibrations curves obtained for compost sample with EC, MM and SA. A Student’s t-test was applied to compare the slope of the calibration curves: EC vs. MM; EC vs. SA; and MM vs. SA. The absence of matrix effect corresponds to slope EC =slope MM; slope EC =slope SA; or slope MM =slope SA. This must be checked for statistical significance. The t calculated is compared with the two-tailed tabulated value, t tab for the appropriate number of degrees of freedom at P% confidence. Typical values are k =2 for 95 % confidence, so, if t <k, the ratio of the slopes is not significantly different from 1, and, if t >k, the ratio of the slopes is significantly different from 1 and the standard addition calibration must be used. The calculated Student’s t (2.027–22.435) (Table S3) showed statistical differences among the slope values of the calibration curves and, consequently, the use of standard addition calibration was required to minimize matrix effects as these interfere with analyte measurement signals. As can be seen, calculated Student’s t increases with the complexity of the matrix obtaining higher values for primary (8.804–12.994), secondary (5.714–21.909) and digested sludge (10.406–22.435). Table 5 Quantification of LAS (mg/kg dw) in soil, commercial fertilizer, compost and different sludge samples by external, matrix-matched and standard addition calibrations. Relative standard deviation percentage (% RSD) is shown in brackets. Soil C10 C11 C12 C13 External calibration 0.52 (5.8) 3.50 (8.3) 3.64 (2.2) 1.94 (2.5) Matrix-matched calibration 0.74 (5.5) 5.07 (3.0) 6.62 (1.4) 3.82 (1.6) Standard addition 0.06 (14) 0.29 (14) 0.36 (8.8) 1.59 (8.3) Commercial fertilizer C10 C11 C12 C13 External calibration 0.93 (9.9) 5.44 (1.9) 5.56 (3.0) 3.25 (0.8) Matrix-matched calibration 1.27 (9.4) 7.38 (1.7) 8.70 (2.1) 5.23 (0.6) Standard addition 1.07 (2.1) 4.93 (4.8) 5.05 (7.8) 5.97 (1.8) Compost C10 C11 C12 C13 External calibration 3.54 (9.4) 19.6 (13) 44.1 (18) 49.6 (17) Matrix-matched calibration 4.59 (9.2) 24.3 (12) 50.3 (17) 55.3 (16) Standard addition 4.29 (11) 33.7 (15) 76.8 (19) 95.9 (18) Digested sludge C10 C11 C12 C13 External calibration 255 (7.6) 1438 (7.1) 2197 (6.2) 1537 (7.1) Matrix-matched calibration 328 (7.5) 1754 (7.0) 2478 (5.9) 1731 (6.8) Standard addition 696 (6.9) 2365 (5.7) 4395 (4.1) 3401 (4.9) Secondary sludge C10 C11 C12 C13 External calibration 17.0 (18) 68.4 (8.8) 120 (8.4) 164 (8.7) Matrix-matched calibration 21.9 (17) 83.6 (8.6) 135 (8.0) 181 (8.6) Standard addition 33.3 (4.4) 135 (4.4) 191 (2.0) 240 (3.2) Primary sludge C10 C11 C12 C13 External calibration 223 (26) 1223 (26) 1606 (34) 1033 (33) Matrix-matched calibration 288 (26) 1497 (26) 1841 (32) 1186 (31) Standard addition 430 (26) 2674 (22) 4120 (30) 3029 (30) J. Martín et al. Microchemical Journal 204 (2024) 111089 7 The external standard calibration has been the most used approach to determine analyte concentrations in analytical procedures [19]. However, its accuracy hinges on the absence of significant interferences in the sample, a condition that is often challenging to meet, particularly in complex samples like sludge. Matrix-matched calibration curves necessitate a similar matrix free from the target compounds, effectively serving as a blank sample. Unfortunately, for substances like LAS, finding a suitable blank matrix is nearly impossible due to its ubiquitous presence in sewage sludge. Moreover, since LAS homologues are generally present in soil, due to historic use of sludge-amendment, and compost samples, due to the frequent addition of sludge in composting, the standard addition method is desirable. Various analytical approaches have been reported for assessing LAS concentrations in sludge samples, with the prevalent use of EC. In contrast, Berg´ e et al. (2016) [20] introduced a distinctive methodology that employs synthetic matrices (MM) for LAS quantification. Our results reveal that the matrix effect has the potential to alter concentration outcomes when employing EC. Simultaneously, it is worth noting that MM, while employed, may also introduce errors owing to the inherent dissimilarity between the actual matrices and the synthetic counterparts, which are not a perfect match. 3.2.2. Method validation The optimized method was validated according to Eurachem Guide [21]. Instrumental limits of quantification (LOQ) were lower than 1 ng/ mL for the commercial LAS sample (calculated from the signal-to-noise ratio =10 injecting solvent fortified with decreasing amounts of LAS). Limits of detection (LODs) were calculated from the signal-to-noise ratio =3 injecting solvents fortified with decreasing amounts of LAS homologues. The method quantification limits (MQLs), and method detection limits (MDLs) were calculated based on the LOQs and LODs, respectively, sample weight, dilution factor of each type of sample, and mean recoveries of LAS at low concentration level (Table 2). The limit of quantification enables the measurement of LAS concentrations up to 1000 times lower than the concentration limits stipulated in the EU directive draft. The concentration range of calibration curves applied to the sludge extracts fulfilled all the statistical requirements for a linear method (Table S2). For all compounds, a well-defined working range was obtained, being characterized by R 2 ≥0.996 using solvent, R 2 ≥0.994 matrix-matched with commercial fertilizer and soil standard addition method, R 2 ≥0.922 with compost standard addition method, R 2 ≥0.956 with secondary sludge standard addition method, R 2 ≥0.991 with digested sludge standard addition method and R 2 ≥0.985 with primary sludge standard addition method. Linear range for LAS homologues varied from 0.1 to 100 mg/kg dw for soil; from 5 to 800 mg/kg dw for compost; from 10 to 1600 mg/kg dw for secondary sludge and from 200 to 240000 mg/kg dw for primary and digested sludge. The accuracy of the proposed method was validated in terms of trueness and precision. Fortified sludge, soil, and compost samples at three concentrations levels (low, medium, and high) were analyzed (Table 3). Precision, expressed as relative standard deviation (%, RSD) was determined from triplicate spiked samples during different days, and the trueness was evaluated using the recovery data by comparing the found vs. spiked level data. The accuracy of the method was satisfactory for all the analytes with relative recoveries values ranging from 84.8 % to 139 % and with an RSD <26 % and <7 % at low and high level, respectively. Data indicate that the methodology to determine LAS in sludge samples using SA calibration is accurate, and that the presence of co-extracted matrix components, which could suppress/enhance the analyte signal in mass spectrometry, did not affect the performance of the method. The specificity of the method was determined by comparing the chromatograms of blank with the corresponding spiked sample. No interferences from endogenous substances were observed at the retention time of the analytes eluted. These finds suggest that the spectrometric conditions ensured high selectivity of the LC–MS/MS method (Fig. S2 and S3). Table 4 presents a concise overview of analytical approaches previously documented for assessing LAS in sludge. Reported methods did not performed any clean-up step while in the proposed methodology the extraction and clean-up were done simultaneously, which is mandatory to decrease matrix effect. Notably, the Soxhlet method [22] demanded the lengthiest processing time (8 h), while pressurized-liquid extraction [23] required the largest sample quantity (10 g). MeOH emerged as the solvent of choice for LAS extraction, although the QuEChERS method [20] necessitated the use of ACN. The proposed methodology utilized the smallest sample size among reported methods and employed extraction times comparable to or shorter than those found in the literature. This is particularly notable given that the majority of the methods cited in the literature did not incorporate a clean-up step. Microwave-assisted extraction [24] methodology employed the lowest solvent volume. Most of the methodologies developed used external calibration for quantification. Obtained recoveries were similar to those obtained in literature. Regarding MQLs, values obtained in this work (0.02–2.89 mg/kg dw) were similar to those methods that employs MS detector (0.025 mg/kg dw [23]) and lower than those obtained with UV/FLD: 0.85–4.40 mg/kg dw [24], 5–56 mg/kg dw [26] or 2.78–15.22 mg/kg dw [27]. 3.3. Method application Method was applied to soil, commercial fertilizer, compost, primary, secondary, and digested sludge using EC, MM and SA as calibration methods (n =5). Results are shown in Table 5. Notably, primary and digested sludge exhibited the highest concentrations, whereas soil displayed the lowest concentration, attributable to a dilution effect. Table 5 reveals that C13 concentrations in soil are quite similar when employing EC and SA calibration methods (1.94 and 1.59 mg/kg dw, respectively). However, in the case of primary sludge, C13 concentrations using SA are three times higher than those obtained using EC (1033 and 3029 mg/kg dw for EC and SA, respectively). These findings underscore the considerable impact of the chosen calibration method on the quantification of LAS in soil, compost, and sludge samples, especially in more complex samples like primary, secondary, and digested sludge. This highlights the importance of selecting an appropriate calibration method to ensure accurate quantification. It is worth noting that the total concentration of LAS in treated sludge such as compost (210.69 mg/kg dw using SA calibration) does not exceed the limits outlined in the EU directive draft for land application of sludge (2,600 mg/kg). 4. Conclusions A method for the determination of LAS homologues C10-C13 in primary, secondary, digested sludge, compost, and soil was optimized and validated. Sample treatment is based on easy-to-perform, costeffective and user-friendly procedure that combine the extraction and clean-up in a single step (UAE-dSPE). The method simplifies both extraction and clean-up, condensing them into a single step while reducing matrix effects through a dilution strategy, aligning with the principles of Green Analytical Chemistry. Analytical determination is carried out by LC–MS/MS. The influence of different matrices resulted a significant feature in the method’s validation. The method of standard addition is the quantitative analysis procedure selected to minimize matrix effects that can interfere with analyte measurement signals. It is key for quality control, in terms of monitoring purposes since matrix effect may introduce a bias in the calculated concentration. The method accuracy in terms of relative recoveries was higher than 84.5 % in all kinds of sludge samples. The limits of quantification allow the measurement of LAS concentrations 1000 times lower than the concentration limits fixed by the EU directive draft for land application of sludge J. Martín et al. Microchemical Journal 204 (2024) 111089 8 (2,600 mg/kg). The proposed method constitutes a useful and accurate tool to evaluate LAS distribution along sludge stabilization treatments. CRediT authorship contribution statement Julia Martín: Writing – review & editing, Methodology, Conceptualization. Carmen Mejías: Visualization, Investigation. Juan Luis Santos: Validation, Methodology, Formal analysis. Irene Aparicio: Writing – review & editing, Validation, Resources. Esteban Alonso: Supervision, Project administration, Funding acquisition, Conceptualization. John Heinze: Project administration, Visualization. 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. Data availability Data will be made available on request. Acknowledgments Authors acknowledge the financial support from the Council for LAB/LAS Environmental Research (CLER). C. Mejías gratefully thanks University of Seville for her predoctoral contract (grant number: VI PPIT-US 2021 II.2A). Appendix A. Supplementary data Supplementary data to this article can be found online at https://doi. org/10.1016/j.microc.2024.111089. References [1] European Union. Working document on sludge, third draft, 27 April 2000. European Union, Brussels. https://www.isprambiente.gov.it/it/progetti/cartellaprogetti-in-corso/suolo-e-territorio-1/uso-dei-fanghi-di-depurazione-in- agricoltura-attivita-di-controllo-e-vigilanza-del-territorio/files/3rd_Draft_sludge_ en.pdf. [2] C. Mejías, J. Martín, J.L. Santos, I. Aparicio, E. Alonso, Occurrence of Pharmaceuticals and Their Metabolites in Sewage Sludge and Soil: A Review on Their Distribution and Environmental Risk Assessment, Trends Environ. Anal. Chem. 30 (2021) e00125, https://doi.org/10.1016/j.teac.2021.e00125. [3] M.M. Gonz´ alez, J. Martín, D. Camacho-Mu˜ noz, J.L. Santos, I. Aparicio, E. Alonso, Degradation and environmental risk of surfactants after the application of compost sludge to the soil, Waste Manag. 32 (7) (2012) 1324–1331, https://doi.org/ 10.1016/j.wasman.2012.02.023. [4] Directive 86/278/EEC on the protection of the environment, and in particular of the soil, when sewage sludge is used in agriculture: Using sewage sludge in farming. [5] HERA Project, Human and Environmental Risk Assessment on ingredients of Household Cleaning Products. LAS, Linear Alkylbenezene Sulphonate (CAS No. 68411-30-3). Revised HERA Report, April 2013. https://www.heraproject.com/ files/HERA-LAS revised April 2013 Final1.pdf. [6] G. Mininni, A.R. Blanch, F. Lucena, S. Berselli, EU policy on sewage sludge utilization and perspectives on new approaches of sludge management, Environ. Sci. Pollut. Res. 22 (2015) 7361–7374, https://doi.org/10.1007/s11356-014-3132- 0. [7] Y. Yu, S. Wang, P. Yu, D. Wang, B. Hu, P. Zheng, M. Zhang, A bibliometric analysis of emerging contaminants (ECs) (2001–2021): Evolution of hotspots and research trends, Sci. Total Environ. 907 (2024) 168116, https://doi.org/10.1016/j. scitotenv.2023.168116. [8] World Health Organization, 1996. Programme, United Nations Environment. Linear Alkylbenzene Sulfonates & Related Compounds (Environmental Health Criteria Series, Volume 169). ISBN 9789241571692. [9] M. Villagrasa, M. Guillamon, E. Eljarrat, D. Barcelo, Matrix effect in liquid chromatography–electrospray ionization mass spectrometry analysis of benzoxazinoid derivatives in plant material, J. Chromatogr. a. 1157 (1,2) (2007) 108–114, https://doi.org/10.1016/j.chroma.2007.04.040. [10] A. Cappiello, G. Famiglini, P. Palma, E. Pierini, V. Termopoli, H. Trufelli, Overcoming matrix effects in liquid chromatography-mass spectrometry, Anal. Chem. 80 (2008) 9343–9348, https://doi.org/10.1021/ac8018312. [11] D. Steiner, R. Krska, A. Malachov´ a, I. Taschl, M. Sulyok, Evaluation of matrix effects and extraction efficiencies of LC-MS/MS methods as the essential part for proper validation of multiclass contaminants in complex feed, J. Agric. Food Chem. 68 (2020) 3868–3880, https://doi.org/10.1021/acs.jafc.9b07706. [12] F. Raposo, D. Barcelo, Challenges and strategies of matrix effects using chromatography-mass spectrometry: An overview from research versus regulatory viewpoints, Trends Anal. Chem. 134 (2021) 116068, https://doi.org/10.1016/j. trac.2020.116068. [13] E. Chambers, D.M. Wagrowski-Diehl, Z. Lu, J.R. Mazzeo, Systematic and comprehensive strategy for reducing matrix effects in LC/MS/MS analyses. J. Chromatogr., B Anal. Technol. Biomed, Life Sci. 852 (2007) 22–34, https://doi. org/10.1016/j.jchromb.2006.12.030. [14] C. Ferrer, A. Lozano, A. Agüera, A.J. Gir´ on, A.R. Fern´ andez-Alba, Overcoming matrix effects using the dilution approach in multiresidue methods for fruits and vegetables, J. Chromatogr. a. 1218 (2011) 7634–7639, https://doi.org/10.1016/j. chroma.2011.07.033. [15] S.L.R. Ellison, M. Thompson, Standard additions: myth and reality, Analyst 133 (2008) 992–997, https://doi.org/10.1039/b717660k. [16] 2002/657/EC: Commission Decision of 12 August 2002 implementing Council Directive 96/23/EC concerning the performance of analytical methods and the interpretation of results. [17] G.A. Gonz´ alez, M.A. Herrador, A practical guide to analytical method validation, including measurement uncertainty and accuracy profiles, Trends Anal. Chem. 26 (3) (2007) 227–238, https://doi.org/10.1016/j.trac.2007.01.009. [18] C. Mejías, M. Arenas, J. Martín, J.L. Santos, I. Aparicio, E. Alonso, Green Assessment of Analytical Procedures for the Determination of Pharmaceuticals in Sewage Sludge and Soil, Crit. Rev. Anal. Chem. (2023), https://doi.org/10.1080/ 10408347.2023.2276294. [19] J.T. Sloop, S.C. Allen, J.A. Carter, G.L. Donati, B.T. Jones, Matrix-matched twopoint calibration based on the standard dilution analysis method, Microchem. J. 160B (2021) 105740, https://doi.org/10.1016/j.microc.2020.105740. [20] A. Berg´ e, B. Giroud, L. Wiest, B. Domenjoud, A. Gonzalez-Espinosa, E. Vulliet, Development of a multiple-class analytical method based on the use of synthetic matrices for the simultaneous determination of commonly used commercial surfactants in wastewater by liquid chromatography-tandem mass spectrometry, J. Chromatogr. a. 1450 (2016) 64–75, https://doi.org/10.1016/j. chroma.2016.04.078. [21] Eurachem. The Fitness for Purpose of Analytical Methods: A Laboratory Guide to Method Validation and Related Topics: Second edition (2014). [22] M.T. García, E. Campos, J. S´ anchez-Leal, I. Ribosa, Effect of linear alkylbenzene sulphonates (LAS) on the anaerobic digestion of sewage sludge, Water Res. 40 (2006) 2958, https://doi.org/10.1016/j.watres.2006.05.033. [23] P. Eichhorn, O. L´ opez, D. Barcel´ o, Application of liquid chromatography–electrospray-tandem mass spectrometry for the identification and characterisation of linear alkylbenzene sulfonates and sulfophenyl carboxylates in sludge-amended soils, J. Chromatogr. a. 1067 (2005) 171, https://doi.org/ 10.1016/j.chroma.2005.01.014. [24] M. Villar, M. Callej´ on, J.C. Jim´ enez, E. Alonso, Optimization and validation of a new method for analysis of linear alkylbenzene sulfonates in sewage sludge by liquid chromatography after microwave-assisted extraction, Anal. Chim. Acta. 599 (2007) 92, https://doi.org/10.1016/j.aca.2007.07.065. [25] J. Riu, E. Martínez, D. Barcel´ o, A. Ginebreda, L.L. Tirapu, LC–MS determination of linear alkylbenzene sulfonates and their carboxylic degradation products in influent and effluent water samples and sludges from sewage-treatment plants, Fresenius J. Anal. Chem. 371 (2001) 448, https://doi.org/10.1007/ s002160101052. [26] J.L. Santos, I. Aparicio, E. Alonso, A new method for the routine analysis of LAS and PAH in sewage sludge by simultaneous sonication-assisted extraction prior to liquid chromatographic determination, Anal. Chim. Acta 605 (1) (2007) 102–109, https://doi.org/10.1016/j.aca.2007.10.027. [27] M. Villar, R. Fernandez-Torres, M. Callej´ on, P. Villar, J.C. Jim´ enez, Application of probe sonication extraction for the determination of linear alkylbenzene sulfonates from sewage sludge. Comparison with other extraction methods, Microchem. J. 90 (2) (2008) 164–170, https://doi.org/10.1016/j.microc.2008.05.005. [28] S. Cantarero, A. Zafra-Gomez, O. Ballesteros, A. Navalon, J.L. Vilchez, C. Verge, J. A. De Ferrer, Matrix effect study in the determination of linear alkylbenzene sulfonates in sewage sludge samples, Environ. Toxicol. Chem. 30 (4) (2011) 813–818, https://doi.org/10.1002/etc.447. J. Martín et al.