Development and Validation of a Method for the Analysis of Multiple Pesticides in Fishery Products Using Gas Chromatography with Micro-Electron Capture Detection and Gas Chromatography–Tandem Mass Spectrometry

Abstract

This study aims to develop a simultaneous analytical method for detecting 19 pesticides, including 4,4′-DDD, in fishery products using gas chromatography with micro-electron capture detection (GC-μECD) and gas chromatography–tandem mass spectrometry (GC-MS/MS). A new analytical method was developed to measure pesticide residues in fishery products based on the modified Association of Official Analytical Chemists protocol combining quick, easy, cheap, effective, rugged, and safe (QuEChERS) and the Pesticide Analytical Manual for extraction and purification. Extraction was performed using acetonitrile containing 0.1% acetic acid, and purification was conducted with Florisil cartridges. The Florisil cartridges were more effective than QuEChERS in removing impurities and pigments during purification and also resulted in a reduced matrix effect. The validation followed Codex guidelines (CAC/GL 40). The limit of detection ranged from 2 to 3 ng/g, and the limit of quantification (LOQ) from 7 to 10 ng/g. Matrix-matched calibration curves exhibited linearity with coefficients of determination exceeding 0.99 for all target analytes. Accuracy was assessed based on recovery rates, while precision was evaluated using relative standard deviations (RSD) at three spiking levels (LOQ, 10×LOQ, and 50×LOQ). The recovery rates ranged from 62.6 to 119.1%, with RSDs of 0.4 to 19.5%, conforming to Codex guidelines.

1. Introduction

Pesticides control weeds, pests, and diseases that harm crops and regulate physiological functions. They provide various benefits, such as improving the quality of agricultural products, increasing yields, reducing costs, and minimizing labor requirements [1]. However, consuming agricultural products or food with inadequately managed pesticide residues may have adverse effects, as most pesticides are toxic synthetic organic compounds. Additionally, when powdered or granular pesticides are directly applied to soil, they are prone to runoff during rainfall due to improper disposal, solubility, and the octanol-water partition coefficient (log K_ow). These pesticides eventually cause water contamination in rivers and seas, leading to bioaccumulation in aquatic organisms like fish [2,3,4,5,6]. This accumulation may cause reproductive failure, ultimately causing ecological disruptions [7,8]. Specifically, organochlorine pesticides damage the immune system, cause congenital abnormalities, impair reproduction, disrupt endocrine functions, and induce cancer. Consequently, their use has been globally restricted since the early 1970s [9]. However, due to environmental stability, long half-lives, and high log K_ow values, these compounds accumulate in the muscles and fat tissues of aquatic organisms and may cause chronic toxicity in humans through prolonged human exposure [10,11].

Recent studies on pesticides and products in aquatic environments have revealed significant findings. Lee et al. [12] reported pesticide residue levels in six major rivers in Korea, detecting eight pesticides, including isoprothiolane, hexaconazole, diazinon, chlorpyrifos, prothiofos, alachlor, butachlor, and molinate. Among these, isoprothiolane, which is widely used in rice cultivation, exhibited the highest detection frequency at 67.5%. Suzuki et al. [13] analyzed pesticide residues in river water near golf courses in Japan and found 12 pesticides, including isoprothiolane. Choi et al. [14] investigated the contamination levels of organochlorine pesticides in edible fish species (olive flounder, rockfish, filefish, and black scraper) near Yongho Wharf in Busan, Korea. Their findings showed residue levels of DDT and HCH ranging from 4.53 to 82.10 µg/L and 2.94 to 22.98 µg/L, respectively. Macgregor et al. [15] reported similar findings, detecting DDT, hexachlorobenzene, and lindane in eels collected from 30 agricultural regions, including Aberdeen, Scotland, through an analysis of persistent organic pollutants. Sharifiarab et al. [16] also noted increasing detection rates of contaminants such as polycyclic aromatic hydrocarbons, polychlorinated biphenyls, and other compounds in aquatic environments. These findings suggest that the pesticides used in agricultural fields and golf courses can contaminate aquatic ecosystems, polluting aquatic organisms. Therefore, continuous monitoring studies are essential to ensure the safe management of aquatic products.

Currently, various analytical methods for pesticides are actively being studied. The Mills method, developed in the United States in the 1960s, is widely used to analyze organochlorine and nonpolar pesticide residues in nonfatty foods. It involves extracting residues with water and acetonitrile (ACN) and partitioning with ether [17]. In the 1970s, pesticide analysis advancements built upon the Mills method, with acetone as an extraction solvent during sample preparation. Gas chromatography detected a broader range of pesticides [18,19]. In the 1980s, efforts focused on reducing the indiscriminate use of solvents in pesticide extraction. Instead of nonpolar solvents, anhydrous MgSO₄ and NaCl were introduced to mitigate environmental pollution and minimize adverse effects on human health [20,21]. Since then, various analytical methods, such as microwave-assisted extraction, accelerated solvent extraction, matrix solid-phase dispersion, and supercritical fluid extraction, have been developed to extract nonpolar pesticides from high-fat samples. However, research on effective impurity removal methods for aquatic product samples remains limited [22,23,24,25,26].

This study focuses on eel, olive flounder, shrimp, and abalone, which are commonly distributed in the market, to compare and analyze the impurity removal efficiency of the modified Association of Official Analytical Chemists (AOAC) and Pesticide Analytical Manual (PAM) purification methods for 19 pesticides, including 4,4′-DDD. The comparison was conducted using total ion chromatograms (TICs) and matrix effect (ME) evaluations.

2. Materials and Methods

2.1. Chemicals and Materials

Nineteen pesticides, including 4,4′-DDD, were purchased from Kemidas (Gunpo, Republic of Korea). ACN and hexane were obtained from J.T. Baker (Center Valley, PA, USA), while dichloromethane (DCM) was sourced from Honeywell (Charlotte, NC, USA), all in HPLC-grade quality. Centrifuge tubes (50 mL) were acquired from SPL (Pocheon, Republic of Korea), and a Thermo Fisher Scientific centrifuge (Waltham, MA, USA) was used. Acetic acid (≥99.7%) and MgSO₄ (99.5%) were obtained from Sigma-Aldrich (St. Louis, MO, USA), while NaOAc (98.5%) was purchased from Junsei (Tokyo, Japan). For purification, Sep-Pak Florisil cartridges (500 mg sorbent) were obtained from Waters (Wexford, Ireland).

2.2. Sample Preparation

Representative species were selected based on the Ministry of Food and Drug Safety Guideline [27] to cover different categories of aquatic products: olive flounder (Paralichthys olivaceus) for marine fish, eel (Anguilla japonica) for freshwater fish, whiteleg shrimp (Litopenaeus vannamei) for crustaceans, and abalone (Haliotis discus hannai) for shellfish. Samples were prepared by removing internal organs and inedible parts, such as heads, tails, and shells. Prepared samples were stored at −20 °C until analysis.

2.3. Sample Extraction and Purification

The AOAC dispersive solid-phase extraction (d-SPE) protocol [28], based on QuEChERS methods for multi-residue pesticide analysis, and the PAM cartridge protocol from the FDA [29] were modified. A 5 g homogenized sample was placed in a 50 mL conical tube, and 20 mL of ACN containing 0.1% acetic acid was added. The mixture was shaken at 2000 rpm for 20 min. Then, 4 g of MgSO₄ and 1.5 g of NaOAc were added, followed by shaking at 2000 rpm for 5 min. The mixture was centrifuged at 4000× g at 4 °C for 10 min to separate the ACN and water layers. ACN (4 mL) from the upper layer was transferred to a conical tube containing 600 mg of MgSO₄ and centrifuged again under the same conditions.

Two purification methods were used.

(1)

d-SPE purification: The ACN upper layer (1 mL) was transferred into a 2 mL centrifuge tube containing 150 mg of MgSO₄, 100 mg of PSA, and 100 mg of C₁₈. The mixture was vortexed for 1 min and centrifuged at 2000× g at 4 °C for 5 min. The final supernatant was used as the test solution.

(2)

Cartridge purification: A 2.5 mL ACN solution was concentrated with a rotary evaporator at 30 °C, then resuspended in 2.5 mL hexane. The Florisil cartridge was activated by passing 5 mL of hexane through it at a flow rate of 2–3 drops/s, with the eluate discarded. Subsequently, 2 mL of the sample dissolved in hexane was passed through the cartridge at the same flow rate and collected in a test tube. Eluent (5 mL, 50% DCM, 3.5% ACN, 46.5% hexane) was added 1 mL at a time and collected in the same test tube. The collected solution was concentrated under nitrogen at 40 °C and dissolved in 1 mL of hexane. The solution was vortexed thoroughly and analyzed using gas chromatography with micro-electron capture detection (GC-μECD) and gas chromatography–tandem mass spectrometry (GC-MS/MS).

2.4. Method Validation

The analytical method for pesticide residue determination in aquatic products was validated following Codex Alimentarius guidelines [30]. The validation parameters included linearity, limit of detection (LOD), limit of quantification (LOQ), recovery, and repeatability. A control sample without detectable pesticide residues was used. Linearity was evaluated by constructing matrix-matched calibration curves and calculating the coefficient of determination (R²). LOD and LOQ were determined by analyzing control samples and calculating the signal-to-noise (S/N) ratio of chromatographic peaks. Peaks with an S/N ratio of ≥3 were defined as the LOD, while those with an S/N ratio of ≥10 were defined as the LOQ. The LOD and LOQ values were established based on the properties of each pesticide. Accuracy and precision were evaluated through recovery tests. Control samples were spiked with mixed standard solutions at LOQ, 10×LOQ, and 50×LOQ concentrations, with each level analyzed in five replicates.

2.5. ME

The ME was assessed by measuring signal suppression or enhancement for the target pesticides in the sample matrix. The extent of signal alteration was calculated as the percentage change in signal intensity observed in the sample matrix relative to that in the solvent. The ME was calculated using Equation (1):

ME (%) = (1 − M/S) × 100

(1)

where M is the slope of the matrix-matched calibration curve (from spiked sample extracts), and S is the slope of the calibration curve from standard solutions prepared in ACN or hexane.

2.6. Instrumental Analysis

Two analytical methods, GC-MS/MS and GC-μECD, were employed. GC-MS/MS analysis was performed using an Agilent Technologies 7010B Triple Quadrupole GC/MS system. Data were processed using MassHunter Quantitative Analysis software (Version 10.1.49, Agilent Technologies, Santa Clara, CA, USA). The analytical conditions are detailed in

Table 1. GC-MS/MS analytical conditions.

Parameters	Condition
Column	DB-5MS UI (30 m × 250 µm × 0.25 µm)
Flow rate	1.2 mL/min
Injection volume	1 µL
Injection mode	Split mode (5:1)
Carrier gas	He
Injection temp.	260 °C
Oven temp.	Rate (°C/min)	Value (°C)	Hold time (min)
		60	0.2
	20	180	0
	15	250	3
	20	300	5
MS/MS condition
Ion source	EI
Source temp.	250 °C
Electron energy	70 eV

, and the precursor and product ions for each compound are listed in

Table 2. Experimental conditions for GC-MS/MS analysis in the multiple-reaction monitoring mode.

Pesticide	Retention Time (min)	Precursor Ion (m/z)	Product Ion (m/z)	Collision Energy (eV)
4,4′-DDD	12.011	237	165	35
		235	165	30
4,4′-DDT	12.593	237	165	35
		235	165	35
Alachlor	9.598	237	160	10
		188	160	10
Atrazine	8.538	215	200	45
		215	58	25
Boscalid	16.491	140	112	15
		140	76	35
Carfentrazone-ethyl	12.361	340	312	15
		312	151	30
Chlorpyrifos	10.106	314	258	25
		199	171	20
α-Endosulfan	11.052	241	206	20
		205	170	20
Fenitrothion	9.841	277	260	5
		277	109	20
Heptachlor	9.648	274	237	20
		272	237	20
Heptachlor epoxide (cis)	10.638	353	253	20
		217	182	30
Iprobenfos	9.165	204	121	45
		204	91	10
Mirex	14.826	272	237	20
		272	143	50
		270	235	20
Prometryn	9.616	241	199	5
		241	184	15
Terbutryn	9.787	241	185	5
		185	170	10
Tetraconazole	10.179	336	218	25
		336	204	40
Trifluralin	8.044	306	264	10
		264	206	5
α-HCH (α-BHC)	8.31	217	181	20
		181	145	25
γ-HCH (γ-BHC, Lindane)	8.73	217	181	10
		181	145	20

. GC-μECD analysis was carried out using an Agilent Technologies 7890A system, with the analytical conditions summarized in

Table 3. GC-μECD analytical conditions.

Instrument	Agilent 7890A gas chromatograph equipped with μECD (Agilent Technologies, Santa Clara, CA, USA)
Column	HP-5 30 m length × 250 μm ID × 0.25 μm film thickness
Temperature	Inlet 260 °C Oven 70 °C (2 min) → 5 °C/min → 158 °C (3 min) → 10 °C/min → 185 °C → 15 °C/min → 195 °C → 20 °C/min → 240 °C (5 min) → 30 °C/min → 300 °C (15 min) Detector 300 °C
Flow rate	Carrier gas (N2) 1.2 mL/min
Injection volume	1 μL
Split ratio	5:1

. Since MS/MS analysis using designated MRM transitions does not allow comprehensive impurity profiling, GC-μECD was additionally employed to compare the overall impurity levels present in fishery products.

3. Results

3.1. Purification Efficiency

Figure 1 compares pigment removal efficiency between the d-SPE and cartridge purification methods for the sample extracts. The eel and olive flounder extracts were colorless, with no noticeable differences between the two purification methods. However, the extracts from abalone and shrimp exhibited distinct color variations; those purified using the Florisil cartridge were nearly colorless, whereas those purified via the d-SPE method exhibited a faint yellow color for abalone and a light yellow hue for whiteleg shrimp. This observation indicates that the Florisil cartridge is more effective at removing pigments during purification.

Figure 2 presents chromatograms of blank extracts prepared using the two purification methods, analyzed with GC-μECD. The chromatograms from the Florisil cartridge purification displayed fewer interfering substances than those from the d-SPE method. The d-SPE method, with its more straightforward purification process, is more prone to residual interfering substances [31]. Conversely, the Florisil cartridge exhibited superior adsorption properties for removing interfering substances, particularly polar and weakly essential compounds, leading to more effective impurity removal. Florisil is known for its strong adsorption capacity for polar substances and lipids, making it highly effective in purifying high-fat samples and removing pesticide residues from food matrices [32,33]. Numerous studies have confirmed the reliability of Florisil cartridges in quantifying organochlorine pesticides by adsorbing various interfering substances, including lipids, from samples such as fishery products, meat, and grains [34,35]. Therefore, the Florisil cartridge purification method provides more accurate analytical results than the relative d-SPE method, especially when monitoring pesticide residues in lipid-rich fishery products containing various interfering substances.

3.2. Sample-Specific ME

The ME values for 19 pesticides were calculated for each sample using the d-SPE and Florisil cartridge purification methods. The results are summarized in

Table 4. ME for 19 pesticides in fishery products using d-SPE and Florisil cartridge purification methods.

Pesticides	Matrix Effect (%)
	Abalone		Eel		Olive Flounder		Whiteleg Shrimp
	d-SPE 1	Cartridge 2	d-SPE	Cartridge	d-SPE	Cartridge	d-SPE	Cartridge
4,4′-DDD	17.7	−11.8	10.7	−13.6	17.0	8.3	−19.9	−4.5
4,4′-DDT	34.2	−14.2	24.5	−15.7	11.8	6.8	−17.8	0.0
Alachlor	35.8	−28.1	31.0	−20.9	19.2	18.2	−43.2	−23.2
Atrazine	31.8	−16.6	19.9	−11.5	12.0	9.7	−28.5	−5.7
Boscalid	86.2	−33.3	103.0	−24.2	135.2	78.4	−56.7	−38.6
Carfentrazone-ethyl	10.9	−37.6	7.3	−26.3	34.1	37.1	−45.7	−4.7
Chlorpyrifos	21.0	7.1	24.1	2.0	10.6	8.4	−15.2	−8.9
α-Endosulfan	17.4	6.2	1.8	9.0	17.1	0.1	−17.1	−15.7
Fenitrothion	54.2	−5.4	15.3	−35.3	49.0	52.5	−27.3	−11.9
Heptachlor	24.6	6.9	28.3	1.1	11.5	4.7	−13.6	−11.3
Heptachlor epoxide (cis)	75.7	−5.7	17.8	−7.2	−0.1	−4.8	−19.6	−7.1
Iprobenfos	49.5	−37.4	24.3	−36.9	43.7	30.7	−38.7	−1.3
Mirex	14.5	−8.9	16.1	−8.1	13.5	12.5	−17.8	−4.3
Prometryn	26.8	−14.1	15.9	−30.4	29.9	30.9	−32.5	−5.8
Terbutryn	45.2	76.7	32.3	−18.4	9.4	18.9	−36.8	4.0
Tetraconazole	36.9	−4.3	29.1	−22.9	31.0	26.7	−38.9	−7.6
Trifluralin	20.3	−5.2	11.7	−10.3	13.2	3.4	−16.7	−4.7
α-HCH (α-BHC)	−4.6	−8.5	−5.8	5.6	4.1	−11.3	−12.4	13.7
γ-HCH (γ-BHC, Lindane)	15.4	1.9	10.0	−3.3	3.7	1.0	−11.8	−0.9

¹ Purification with d-SPE. ² Purification with Florisil cartridge.

. The ME was categorized according to the classification system proposed by Chatterjee [36], as follows: high signal suppression (ME > −50%), moderate suppression (−50% ≤ ME < −10%), no suppression or enhancement (−10% ≤ ME ≤ 10%), moderate signal enhancement (10% < ME ≤ 50%), and high enhancement (ME > 50%). For the d-SPE method, the ME values ranged from −4.6 to 86.2% for abalone, −5.8 to 103.0% for eel, −0.1 to 135.2% for olive flounder, and −56.7 to −11.8% for whiteleg shrimp. Conversely, the Florisil cartridge method resulted in ME values ranging from –37.6 to 76.7% for abalone, −36.9 to 9.0% for eel, −11.3 to 78.4% for olive flounder, and –38.6 to 13.7% for whiteleg shrimp. Most of the samples exhibited ME values within the −50 to 50% range. However, certain pesticides, such as boscalid, fenitrothion, and heptachlor epoxide (cis), showed higher signal suppression or enhancement with the d-SPE method than with the Florisil cartridge method.

This variation in ME values can be attributed to the substantial organic content, including proteins and lipids [37,38]. Previous studies have shown that significant amounts of interfering substances are often co-extracted with samples, even after purification, which can affect the analytical results. Moreover, the presence of these interfering substances during extraction can lead to contamination of the GC injection port, resulting in reduced sensitivity and precision in intra- and inter-sequence analyses. These issues may also increase the frequency of instrument maintenance and extend downtime [39]. Such challenges can introduce errors in both qualitative and quantitative results when monitoring fishery products, thereby reducing the reliability of the analysis. Therefore, developing reliable analytical methods to eliminate interfering substances through additional purification or dilution steps is essential.

3.3. LOD, LOQ, and Recovery Rate

The LOD for the developed Florisil cartridge method ranged from 2 to 3 ng/g, while the LOQ was generally between 7 and 9 ng/g, except for the lipophilic pesticide property, which had an LOQ of 10 ng/g (

Table 5. Linearities, LOD, LOQ, recoveries, and precisions of multiclass pesticides.

Pesticides	Matrix	Linearity (R2)	LOD (ng/g Wet Mass)	LOQ (ng/g Wet Mass)	Recovery (%) ± SD 1			RSD 2 (%)
					LOQ	10×LOQ	50×LOQ	LOQ	10×LOQ	50×LOQ
4,4′-DDD	Eel	0.99896	2	7	102.6 ± 1.1	103.4 ± 2.3	107.8 ± 1.1	1.0	2.2	1.0
	Olive flounder	0.99944	2	7	101.0 ± 1.6	102.0 ± 2.4	108.0 ± 4.7	1.5	2.3	4.3
	Avalone	0.99968	2	7	82.6 ± 2.2	78.2 ± 0.8	81.6 ± 5.4	2.6	1.0	6.6
	Whiteleg shrimp	0.99933	2	7	98.8 ± 1.4	98.6 ± 1.7	99.8 ± 3.3	1.4	1.7	3.3
4,4′-DDT	Eel	0.99748	2	7	102.4 ± 1.0	103.0 ± 2.3	107.6 ± 1.2	0.9	2.2	1.1
	Olive flounder	0.99537	3	9	101.4 ± 2.3	104.1 ± 2.2	109.8 ± 5.6	2.2	2.1	5.1
	Avalone	0.99948	2	7	86.0 ± 1.5	81.2 ± 2.5	86.8 ± 5.8	1.7	3.0	6.6
	Whiteleg shrimp	0.99881	2	7	101.6 ± 2.4	102.7 ± 2.3	101.9 ± 4.4	2.3	2.2	4.3
Alachlor	Eel	0.99944	2	7	103.2 ± 1.0	105.0 ± 2.1	110.8 ± 1.2	0.9	2.0	1.0
	Olive flounder	0.99916	2	8	93.2 ± 2.7	92.2 ± 0.9	93.1 ± 3.5	2.8	0.9	3.7
	Avalone	0.99341	2	7	82.1 ± 3.3	78.2 ± 3.9	79.3 ± 2.6	4.0	4.9	3.2
	Whiteleg shrimp	0.99969	2	7	86.9 ± 3.3	81.0 ± 4.7	81.7 ± 4.8	3.7	5.8	5.8
Atrazine	Eel	0.99946	2	7	96.7 ± 1.5	99.0 ± 1.8	103.9 ± 1.9	1.5	1.8	1.8
	Olive flounder	0.99974	2	7	97.4 ± 1.7	94.7 ± 3.4	97.7 ± 3.7	1.7	3.5	3.7
	Avalone	0.99812	2	7	83.3 ± 2.7	76.6 ± 3.5	78.9 ± 2.9	3.2	4.5	3.6
	Whiteleg shrimp	0.99868	2	7	82.9 ± 4.8	83.6 ± 3.1	81.5 ± 5.9	5.7	3.7	7.2
Boscalid	Eel	0.99728	2	7	107.2 ± 1.5	109.1 ± 2.6	116.4 ± 2.0	1.3	2.3	1.7
	Olive flounder	0.99976	2	7	86.0 ± 4.0	69.6 ± 7.1	68.4 ± 6.0	4.6	10.2	8.7
	Avalone	0.99630	2	7	73.3 ± 11.6	72.5 ± 2.9	68.0 ± 11.4	15.8	4.0	16.8
	Whiteleg shrimp	0.99449	2	7	81.8 ± 9.5	72.8 ± 2.8	78.34 ± 2.4	11.6	3.8	3.0
Carfentrazone-ethyl	Eel	0.99640	2	7	104.4 ± 1.2	105.8 ± 1.7	111.2 ± 0.5	1.1	1.6	0.4
	Olive flounder	0.99808	2	7	92.0 ± 5.4	82.5 ± 5.7	84.8 ± 3.6	5.8	6.9	4.2
	Avalone	0.99794	2	7	70.5 ± 14.0	68.7 ± 12.8	67.6 ± 4.9	19.8	18.6	7.2
	Whiteleg shrimp	0.99910	2	7	82.5 ± 8.7	90.4 ± 10.6	91.9 ± 3.7	10.5	11.7	4.0
Chlorpyrifos	Eel	0.99968	2	7	101.9 ± 1.1	104.2 ± 2.4	109.4 ± 1.0	1.0	2.3	0.9
	Olive flounder	0.99948	2	7	91.4 ± 4.0	89.4 ± 6.4	95.2 ± 3.1	4.3	7.1	3.2
	Avalone	0.99931	2	7	84.0 ± 3.3	78.24 ± 1.5	80.79 ± 4.7	3.9	1.9	5.8
	Whiteleg shrimp	0.99857	2	7	87.9 ± 2.7	84.81 ± 3.6	85.06 ± 3.6	3.0	4.2	4.2
α-Endosulfan	Eel	0.99809	2	7	103.5 ± 1.2	105.0 ± 2.2	111.1 ± 0.8	1.1	2.0	0.7
	Olive flounder	0.99891	2	7	89.5 ± 5.5	87.6 ± 5.7	92.6 ± 3.2	6.1	6.5	3.4
	Avalone	0.99825	2	7	83.8 ± 7.7	85.68 ± 6.3	83.48 ± 9.5	9.1	7.3	11.3
	Whiteleg shrimp	0.99890	2	7	90.05 ± 8.3	95.57 ± 4.3	87.11 ± 7.1	9.2	4.4	8.1
Fenitrothion	Eel	0.99936	2	7	103.1 ± 1.5	105.0 ± 2.6	111.3 ± 1.1	1.4	2.4	0.9
	Olive flounder	0.99977	2	7	97.3 ± 2.6	89.8 ± 2.3	93.6 ± 4.1	2.6	2.5	4.3
	Avalone	0.99413	2	7	85.82 ± 1.9	79.21 ± 4.2	83.19 ± 6.5	2.2	5.3	0.7
	Whiteleg shrimp	0.99846	2	7	86.55 ± 2.8	96.93 ± 6.6	96.46 ± 10.2	3.2	6.8	10.5
Heptachlor	Eel	0.99912	2	7	100.8 ± 0.8	103.4 ± 2.5	109.0 ± 1.7	0.7	2.4	1.5
	Olive flounder	0.99895	2	7	94.2 ± 2.8	96.5 ± 2.6	101.2 ± 4.3	2.9	2.6	4.2
	Avalone	0.99896	2	7	87.18 ± 1.0	86.38 ± 2.3	88.38 ± 3.9	1.1	6.1	4.4
	Whiteleg shrimp	0.99943	2	7	98.06 ± 0.6	97.97 ± 2.8	96.21 ± 3.1	0.6	2.8	3.2
Heptachlor epoxide (cis)	Eel	0.99804	2	7	101.5 ± 1.5	104.2 ± 2.9	109.4 ± 1.1	1.4	2.7	1.0
	Olive flounder	0.99891	2	7	98.1 ± 2.6	100.1 ± 4.4	105.2 ± 4.4	6.8	4.3	4.1
	Avalone	0.99472	2	7	83.23 ± 6.7	80.25 ± 3.0	85.7 ± 10.5	8.0	3.7	12.2
	Whiteleg shrimp	0.99808	2	7	98.4 ± 7.0	96.04 ± 5.1	73.91 ± 2.4	0.6	5.3	3.2
Iprobenfos	Eel	0.99172	2	7	103.8 ± 1.2	106.2 ± 2.3	112.3 ± 1.4	1.1	2.1	1.2
	Olive flounder	0.99955	2	7	97.8 ± 1.8	88.7 ± 1.4	89.0 ± 4.3	1.8	1.5	4.8
	Avalone	0.99932	2	7	81.62 ± 2.1	78.88 ± 2.3	78.44 ± 4.9	2.5	2.9	6.2
	Whiteleg shrimp	0.99823	2	7	84.45 ± 4.1	84.08 ± 4.0	83.05 ± 4.1	4.8	4.7	4.9
Mirex	Eel	0.99953	3	9	107.62 ± 1.7	111 ± 2.8	117.63 ± 1.7	1.5	2.5	1.4
	Olive flounder	0.99842	2	7	103.06 ± 2.2	102.6 ± 1.1	108.36 ± 2.6	2.1	1.0	2.3
	Avalone	0.99952	2	7	68.38 ± 1.1	65.89 ± 0.8	68.78 ± 2.8	1.6	1.2	4.0
	Whiteleg shrimp	0.99955	2	7	80.51 ± 0.9	81.72 ± 0.8	80.91 ± 3.1	1.1	0.9	3.8
Prometryn	Eel	0.99880	3	10	106.19 ± 2.2	109.1 ± 3.0	115.9 ± 1.8	2.0	2.7	1.5
	Olive flounder	0.99921	3	10	87.10 ± 1.5	82.75 ± 2.8	87.49 ± 2.5	1.7	3.3	2.8
	Avalone	0.99933	2	8	64.22 ± 3.5	63.77 ± 3.1	62.61 ± 4.9	5.4	4.8	7.8
	Whiteleg shrimp	0.99909	3	9	80.79 ± 2.9	75.48 ± 2.9	76.07 ± 6.7	3.5	3.8	8.8
Terbutryn	Eel	0.99806	2	8	103.78 ± 1.1	109.65 ± 3.5	115.79 ± 1.2	1.0	3.1	1.0
	Olive flounder	0.99791	2	8	99.42 ± 2.1	92.69 ± 5.4	97.86 ± 2.2	2.1	5.8	2.2
	Avalone	0.99788	2	7	85.28 ± 4.0	80.53 ± 2.6	65.45 ± 9.2	4.6	3.2	14
	Whiteleg shrimp	0.99936	2	7	78.26 ± 6.7	75.17 ± 6.5	76.16 ± 4	8.5	8.6	5.2
Tetraconazole	Eel	0.99184	2	7	106.56 ± 1.2	111.0 ± 2.5	117.09 ± 2.2	1.1	2.2	1.8
	Olive flounder	0.99586	2	7	119.09 ± 12.4	100.0 ± 16.7	114.12 ± 12.0	10.4	16.7	10.5
	Avalone	0.99661	2	7	105.46 ± 18.4	96.93 ± 7.9	88.83 ± 12.2	17.4	8.1	13.7
	Whiteleg shrimp	0.99430	2	7	108.99 ± 12.7	94.59 ± 18.5	91.25 ± 8.9	11.6	19.5	9.7
Trifluralin	Eel	0.99093	2	7	109.3 ± 1.5	111.64 ± 2.6	117.94 ± 1.7	1.3	2.3	1.4
	Olive flounder	0.99937	2	7	101.63 ± 2.2	101.14 ± 2.8	104.14 ± 2.7	2.1	2.7	2.5
	Avalone	0.99643	2	7	87.22 ± 2.1	82.96 ± 2.6	85.95 ± 4.1	2.4	3.1	4.7
	Whiteleg shrimp	0.99723	2	7	93.73 ± 2.8	93.27 ± 1.9	93.32 ± 5.0	2.9	2.0	5.3
α-HCH (α-BHC)	Eel	0.99955	2	7	104.26 ± 0.8	104.86 ± 2.6	110.24 ± 1.4	0.7	2.4	1.2
	Olive flounder	0.99946	2	7	101.33 ± 1.5	104.1 ± 2.8	108.45 ± 2.2	1.4	2.6	2.0
	Avalone	0.99908	2	8	101.66 ± 1.4	92.13 ± 5.8	100.98 ± 3.5	1.3	6.2	3.4
	Whiteleg shrimp	0.99841	2	7	94.35 ± 2.7	92.91 ± 0.8	91.44 ± 3.4	2.8	0.8	3.7
γ-HCH (γ-BHC, Lindane)	Eel	0.99976	2	7	103.16 ± 1.3	104.6 ± 2.7	109.75 ± 1.9	1.2	2.5	1.7
	Olive flounder	0.99921	2	7	92.60 ± 2.3	96.0 ± 2.9	99.81 ± 2.9	2.4	3.0	2.9
	Avalone	0.99937	2	7	89.14 ± 2.8	85.51 ± 1.5	87.47 ± 3.6	3.1	1.7	4.1
	Whiteleg shrimp	0.99940	2	7	96.16 ± 2.1	96.84 ± 1.8	94.18 ± 4.2	2.1	1.8	4.4

¹ Standard deviation. ² Relative standard deviation of repeatability.

). Accuracy was determined through recovery tests, and precision was assessed by calculating the relative standard deviation (RSD) for each representative species. The recovery test showed the following results:

• Abalone: 64.2–105.5% (LOQ), 63.7–96.9% (10×LOQ), and 62.6–88.8% (50×LOQ);
• Eel: 96.7–109.3%, 99.0–111.6%, and 103.9–117.9%;
• Olive flounder: 86.0–119.1%, 69.6–104.1%, and 87.5–114.1%;
• Whiteleg shrimp: 78.3–109.0%, 72.8–102.7%, and 76.1–101.9%.
• The RSD values were as follows:
• Abalone: 1.1–19.8%, 1.0–18.6%, and 0.7–16.8%;
• Eel: 0.7–1.5%, 1.6–2.7%, and 0.4–1.7%;
• Olive flounder: 1.4–10.4%, 0.9–16.7%, and 2.0–10.5%;
• Whiteleg shrimp: 0.6–11.6%, 0.8–19.5%, and 3.0–10.5%.

The identical sequences of LOQ, 10×LOQ, and 50×LOQ denote the results. The accuracy and precision satisfied the analytical method validation conditions of the Codex guidelines (spike concentrations: 1–10 ng/g and 10–100 ng/g; recovery rate: 60–120% and 70–120%; RSD: ≤30% and 20%).

3.4. Calibration Curve and Linearity

To assess the linearity of the developed analytical method, matrix-matched calibration solutions were prepared at various concentrations based on the LOQ and injected into the GC-MS/MS system. The concentration ranges for each LOQ were as follows: 1.4, 3.5, 4.9, 7, 8.4, 10.5, and 14 ng/g for an LOQ of 7 ng/g; 1.6, 4, 5.6, 8, 9.6, 12, and 16 ng/g for an LOQ of 8 ng/g; 1.8, 4.5, 6.3, 9, 10.8, 13.5 for an LOQ of 9 ng/g, and 18 ng/g; and 2, 5, 7, 10, 12, 15, and 20 ng/g for an LOQ of 10 ng/g. The R² values for the calibration curves of the 19 pesticides ranged from 0.99093 to 0.99977, satisfying the Codex guideline requirement (R² > 0.98) [30]. This result confirms excellent linearity, demonstrating the suitability of the method for quantitative analysis during monitoring.

4. Conclusions

This study developed an optimized sample preparation and analytical method for detecting pesticide residues in fishery products by modifying the PAM method with a Florisil cartridge purification step. The method was validated using GC-μECD and GC-MS/MS. The results indicated that the Florisil purification method yielded higher recovery rates and lower MEs than the d-SPE method. The validated method exhibited high accuracy (recovery: 62.6–119.1%) and precision (RSD: 0.4–19.5%), with coefficients of determination (R² > 0.98) meeting Codex guideline requirements. These results demonstrate that the developed method offers improved impurity removal during pesticide residue analysis in fishery products, while also reducing the frequency of instrument maintenance. As a result, the method is considered more accurate and operationally efficient than existing approaches. This analytical approach provides critical baseline data for safety evaluation and management of fishery products. Its application in continuous monitoring is expected to enhance the safety and quality assurance of fishery products. However, as the developed analytical method targets only 19 pesticide compounds, future research should focus on expanding the scope to include multi-residue detection and on minimizing the use of reagents and solvents in the pretreatment process to establish more environmentally friendly analytical approaches.