3.1. Nanocomposite characterization
As presented in Fig. 1 magnetic properties of silica aerogel was performed through chemical co-precipitation reaction. Moreover, hydrothermal strategy in the presence of copper sulphate, MSi and IL was used for MSi@CuR@IL production.
Figure 2 specified the SEM, Elemental analysis of C/H/N, VSM, TGA, BET and FTIR character of synthetized nanocomposite. The morphology of prepared MSi@CuR@IL was characterized by SEM technique. Image in Fig. 2a, 2b show the tetrahedral structure shape of CuO nanorods by diameter between 69.88 to 311.3 nm.However nanostructure of silica aerogel was confirmed by the size of 38.88 and 44.67nm.
FT-IR spectra of two synthetize steps for production of MSi and MSi@CuR@IL are depicted in Fig. 2c, 2d. The peaks at 457 cm− 1, 3210 cm− 1 and 3180 cm− 1 are specified in these components. The FTIR band at 457 cm− 1 in Fig. 2c is associated to the Cu-O stretching band [28]. Peaks in the region between 3100 cm− 1 to 3200 cm− 1 in (Fig. 2d) are related to quaternary amine groups of IL component [29]. Moreover bands at 3440 cm− 1 were related to stretching vibrations of N-H groups.
Results of N2 absorption/desorption in BET for MSi@CuR@IL was obtained at 77 K (Fig. 2f ) with the type V isotherm. The slow adsorption at low pressure and hysteresis loop at high level confirmed presence of mesoporous structure. Surface area of nanocomposite was measured 29.6 m2/g by total pores volume of 0.08cm3/g and porosity size of 11.44 nm.
Presence of IL in the MSi@CuR@IL nanocomposite was determined by C/H/N elemental analyser (Fig. 2g). According to C/H/N analysis 0.172% wt. nitrogen, 0.235% wt. C, 1.296% wt. H, and C/N ratio, 1.3 was found in the nanocomposite. EDS analysis also confirmed MSi@CuR@IL synthesis by the existence of C, N, Fe, O and Si (Table 2) in their structure.
Table 2
EDS Analysis of adsorbent
Chemical | Element | Weight % |
MSi@CuR@IL | C | 1.97 |
N | 1.25 |
O | 47.18 |
Cu | 47.71 |
Si | 1.7 |
Fe | 0.19 |
The Magnetic properties of MSi@CuR@IL are presented in Fig. 2h. S- Shape of Magnetic loops by the saturation of 6.2 emu/g was seen. Figure 2e.presented the results of TGA test for MSi@CuR@IL component. Weight loss of about 50% in the TGA curve occurred at 350°C.
3.2. DSPE method optimization
Significant factors such as pH, MSi@CuR@IL amount, and suspension speed, time of absorption, type and volume of elution were optimized in DSPE method. Between methanol, chloroform and acetonitrile solvents, phthalates extraction was optimized by methanol based on the Fig. 3a.
Moreover elution volume is an important parameter for successful extraction process. Large volumes of eluent is time-consuming due to nitrogen drying of extraction and analyte extraction is not completed in the little amount of solvent. Figure 3b shows 2 mL of methanol is sufficient for DEHP, DMP, and DEP extraction.
The mean GC response of DEHP, DMP, DEP chemicals(R) in DSPE method were fitted to X1, X2, X3 and X4 by linear model. Following equation predict the mean of GC response in phthalates to independent variables.
R = 91.03130-0.229781 X1-3.62090 X2 + 1.57520 X3-0.051071 X4
ANOVA analysis of CCD-RSM method presented in Table 3. F-value of 55.22 and the p-value of 0.0001 confirmed significance of linear model. R2 coefficients of 0.94 and adjusted R2 of 0.75 with predicted R2 of 0.89 indicated suitable describing of phthalate extraction.
Table 3
ANOVA analysis of linear model.
Source | Sum of squares | Degree freedom | Mean Square | F-value | p-value Prob > F |
Model | 3813.85 | 4 | 953.46 | 55.22 | 0.0001 |
X1 | 19.92 | 1 | 19.92 | 1.15 | 0.3 |
X2 | 569.75 | 1 | 569.75 | 33 | 0.0001 |
X3 | 673.91 | 1 | 673.91 | 39.03 | 0.0001 |
X4 | 2550.28 | 2550.28 | 147.71 | 12.5 | 0.0001 |
Residual | 224.45 | 13 | 17.27 | | |
Lack of Fit | 176.78 | 11 | 16.07 | 0.6742 | 0.7311 |
Pure Error | 47.67 | 2 | 23.84 | | |
Cor Total | 4038.30 | 17 | | | |
The interaction between X2 (pH) with X3 (time of sample absorption) and X4 (suspension speed) represent higher phthalate extraction at lower pH. At down pH the functional groups of MSi@CuR@IL are increased due to protonation and production of positive charge in the surface of nanocomposite. In fact generation of interaction between phthalate polar site and MSi@CuR@IL is accelerate. However extraction potency increased by growing the time of sample absorption [30]. According to linear model, the best extraction was performed using 24 mg MSi@CuR@IL, sample absorption time of 14 min at pH = 6.2 by suspension speed of 700 RPMI.
3.3. Method Evaluation
Adsorbent potency of MSi, MSi@CuR and MSi@CuR@IL was compared in optimized DSPE method. The growth of extraction level presented in Fig. 5.
Phthalate chemicals has the reaction sites such as carbonyl group for electrostatic interaction and hydrogen bonding and the site of aromatic and alkyl group for van der Waals bonding. In this way phthalates could be interact with polar component by their carbonyl groups and nonpolar forces is created by aromatic and alkyl groups. The increase of polar groups from MSi@CuR than MSi developed the adsorption potency at one step and in the second stage polar and nonpolar sites of IL progress DEHP, DMP, DEP extraction in DSPE method with MSi@CuR@IL adsorbent.
To evaluate performance of optimized method, parameters of LOD, LOQ, intera-day and inter-day reproducibility (n = 3) were measured (Table 4).
Table 4
Validation parameters of the proposed method.
Sample | Calibration curve equation | R2 | LOD (µg/L) | LOQ(µg/L) | Inter-day RSD% | Intera-day RSD% |
5 | 60 (µg/L) | 200 | 5 | 60 (µg/L) | 200 |
DMP | y = 0.093x + 0.6035 | 0.993 | 0.78 | 2.57 | 9.5 | 7 | 3.3 | 9.5 | 3.7 | 2.9 |
DEP | y = 0.008x + 0.2241 | 0.991 | 0.84 | 2.77 | 9.4 | 8.1 | 3.5 | 6.4 | 1.7 | 2.1 |
DEHP | y = 0.0151x + 4.3479 | 0.999 | 1.25 | 4.12 | 6.4 | 4.5 | 4.5 | 3.3 | 4.8 | 4.4 |
Calibration plots was performed in the range of 5 to 200 µg/L with regression ranging from 0.991 to 0.999. Extraction recovery for DEHP, DMP, and DEP was evaluated 100.22 ± 4.3, 92.21 ± 9.5 and 102.12 ± 4.5 respectively. The intra-day and inter-day %RSD as the precision of method were reported between 1.7 and 9.5. These results confirmed that DSPE containing MSi@CuR@IL nanocomposite is suitable for phthalate assessment in water samples. The results of phthalate contamination of plastic bottled water are presented in Table 5.
Table 5
DEHP, DMP, and DEP assessment in plastic bottled water samples
Sample number | DMP (ppb) | DEP(ppb) | DEHP(ppb) |
1 | 90.4 | 74.8 | 74.8 |
2 | 44 | 125.4 | 125.4 |
3 | 91.4 | 115.8 | 115.8 |
4 | 79 | 170 | 170 |
5 | 171 | 138.6 | 138.6 |
6 | 106 | 170 | 170 |
7 | 119.3 | 188 | 188 |
8 | 54.3 | 195.3 | 195.3 |
9 | 8.9 | 187 | 187 |
10 | 34.7 | 161.5 | 161.5 |
We compare DSPE based MSi@CuR@IL with reported methods for phthalate in Table 6. The results show validation parameters of suggested method are competitive with other reports.
Table 6
Comparison between DSPE based MSi@CuR@IL and other reported methods for DEHP, DMP, and DEP assessment
Technique | Analyte | Sample Pretreatment | Matrix | Linear range (µg/L) | Limit of detection (µg/L) | RSD (%) | Reference |
GC-FID | Phthalate Easers | solid-phase extraction | Beverages | 5-750 | 0.85–1.38 | 2.54–6.33 | [31] |
GC-FID | DOP – DIBP-DEHA – DBP | dispersive micro solid phase extraction | Aqueous samples | - | 0.88 − 1.04 | - | [32] |
GC-FID | DMP-DEP-DEHP-DBP-BBP | Solid-phase microextraction | cosmeceutical products | 5-1000 | 3.65–4.91 | < 5.5 | [33] |
GC-FID | DMP-DEP -DBP-BBP-DNOP | Solid-phase microextraction | aqueous samples | 5-500 | 0.32–3.13 | 3.7–13.3 | [34] |
GC-FID | DMP-DEP-DEHP | dispersive solid phase extraction | Aqueous samples | 5-200 | 0.78–1.25 | 1.7–9.5 | This study |
Based on the importance of phthalate release from plastic bottled water, we optimized the novel DSPE based on the new adsorbent in an accurate and repeatable method. However, Gas chromatography with electron capture detector (GC-ECD) was used in method 606 EPA followed by liquid-liquid extraction. In our study, gas chromatography with flame ionization detector (GC-FID) was suggested. FID is common detector and user friendly while ECD is expert dependent with uncertainty of multiple negative ions [30]. Although ECD determine has higher sensitivity than FID, we optimized the GC-FID based method by suitable LOD due to appropriate sample preparation with MSi@CuR@IL nanocomposite.