Water pollution from dyes is a significant global concern, impacting access to clean water and posing health risks. Over 7 × 10⁵ tons of dyes are produced annually, with 15% ending up in wastewater. Dyes are highly soluble, persistent, and toxic, affecting aquatic life and human health through allergic reactions, asthma, DNA damage, and cancer. Many dyes are not biodegradable. Accurate and reliable methods for dye detection and removal are crucial. Existing methods like capillary electrophoresis, thin layer chromatography, HPLC, electrochemistry, voltammetry, and spectrophotometry have limitations. Spectrophotometric methods are widely used due to their simplicity and cost-effectiveness. However, sample preparation is essential. Liquid–liquid extraction (LLE) is less common now due to its high solvent consumption and long extraction times. Dispersive liquid–liquid microextraction (DLLME) offers advantages with its simplicity, speed, and reduced solvent use. Ultrasound-assisted DLLME (UA-DLLME) further enhances efficiency by using cavitation to increase mass transfer and reduce extraction time. Response surface methodology (RSM) is a statistical method for efficient optimization of extraction parameters. This study aims to optimize UA-DLLME for the extraction of RB and MG dyes from aqueous solutions using RSM, specifically a central composite design (CCD), to determine the optimal conditions for maximum extraction efficiency.

The literature review section discusses various existing methods for determining dyes, highlighting their advantages and disadvantages. It emphasizes the limitations of traditional LLE and the advantages of DLLME and UA-DLLME techniques for improving extraction efficiency and reducing solvent usage. Several studies on the extraction and determination of malachite green and Rhodamine B using different techniques such as HPLC, spectrophotometry, and other microextraction methods are reviewed, setting the stage for the proposed UA-DLLME method and its optimization using RSM.

The study employed UA-DLLME coupled with UV/Vis spectrophotometry for the determination of RB and MG dyes. Different extraction and disperser solvents were tested (chloroform, trichloromethane, methanol, ethanol, etc.), with chloroform and ethanol selected as optimal. The effect of centrifuge speed (1000–4500 rpm) was investigated, with 3500 rpm selected as the optimal speed. RSM with a CCD was used to optimize the following parameters: pH (3–7), sonication time (1–5 min), volume of extraction solvent (chloroform, 50–250 µL), and volume of disperser solvent (ethanol, 200–1000 µL). The effect of salt addition was also studied. A quadratic polynomial equation was used to model the relationship between extraction efficiency and the variables. ANOVA was performed to assess the model's fitness. The model's adequacy was evaluated using R², Adj-R², and lack-of-fit test. Residual plots were examined for normality and independence of errors. Three-dimensional response surface plots were generated to visualize the interactions between variables. Optimization was performed using Design Expert software to determine the conditions for maximum extraction efficiency. Method validation included determination of the linear dynamic range (LDR), limit of detection (LOD), limit of quantification (LOQ), preconcentration factor (PF), enrichment factor (EF), extraction recovery (%ER), and relative standard deviation (%RSD). The method was applied to real water samples (deionized water, tap water, lake water, and wastewater) to assess its applicability. The results were compared with those obtained using other methods reported in the literature.

The optimal conditions for maximum dye extraction were found to be: pH 5, 4 min sonication time, 120 µL of chloroform as extraction solvent, and 760 µL of ethanol as disperser solvent. No salt addition was needed. Under these conditions, the maximum experimental extraction efficiencies were 97.68% for RB and 98.51% for MG. The method exhibited a wide linear dynamic range (7.5–1500 ng mL⁻¹ for RB and 12–1000 ng mL⁻¹ for MG), low limits of detection (1.45 ng mL⁻¹ for RB and 2.73 ng mL⁻¹ for MG), and good precision (RSD < 3.5%). The preconcentration factor was 83.33 for both dyes. The developed method successfully determined RB and MG in real water samples with recoveries ranging from 95.53% to 99.60%. Comparison with other methods demonstrated that the UA-DLLME method offers comparable or superior performance in terms of detection limits and linear range.

The findings demonstrate the effectiveness of UA-DLLME coupled with RSM for efficient and sensitive determination of RB and MG dyes in water samples. The optimized method provides a simple, cost-effective, and rapid alternative to existing techniques. The high extraction efficiency and low detection limits make it suitable for monitoring dye contamination in various water sources. The use of RSM significantly reduced the number of experiments required for optimization, highlighting the efficiency of the statistical approach. The successful application to real water samples validates the method’s practicality and robustness for environmental monitoring.

This study successfully developed and optimized an ultrasound-assisted dispersive liquid–liquid microextraction (UA-DLLME) method coupled with UV-Vis spectrophotometry for the simultaneous determination of malachite green and rhodamine B in water samples. The method offers advantages such as simplicity, low cost, high speed, high efficiency, and low detection limits. The use of response surface methodology significantly enhanced the optimization process. Future research could explore the application of this method to other types of dyes and matrices, and investigate the use of alternative solvents for improved sustainability.

While the method demonstrated high efficiency and sensitivity, potential limitations include the possible matrix effects from complex samples, which might require further sample preparation steps. The study focused on two specific dyes; further investigation is needed to assess the method's applicability to a broader range of dyes with varying chemical properties. The optimal conditions determined in this study might not be universally applicable and may need to be adjusted depending on the specific sample characteristics.