3.1. Characterization of electrospun nanofibers
The field emission- scanning electron microscopy (FE-SEM) images are exhibited in Fig. 1. As has been illustrated, Fig. 1.A. and Fig. 1.B. demonstrate the FE-SEM images of the nanofibers in different magnifications. As can be seen, the fibers have a non-bead structure and are oriented randomly. The polymeric fibers sizes are specified on the FE-SEM images. It is obvious that the nanofibers do not have the same diameters and the average diameters are in the range of 111–138 nm. The EDS analysis of the fabricated nanofibers was investigated and the results are shown in Fig. 2.A. The EDS analysis represents the peaks of carbon, nitrogen, oxygen, and chlorine atoms with molar ratios (W%) of 49.6%, 1.9%, 6.1%, and 42.4%, respectively. The elemental mapping analysis (Fig. 2.B) also proves the distribution of the mentioned atoms on the surface of the nanofibers.
(ATR) - (FTIR) spectra of PVC/EVA/CS/Str nanofibers in the range of 4000 cm-1 to 670 cm-1, are shown in Fig S.2. Both adsorption peaks at 2911.4 cm-1 and 2968.3 cm-1, are related to CH2 and C-H symmetric stretching vibrations in PVC, respectively. The peaks at 1425 cm-1 and 1328 cm-1 are attributed to C-H bending vibrations in PVC. The wavenumbers between 966.3 to 1248.7 cm-1 are contributed to C-C stretching in PVC. The bond at 3350 cm-1 is related to -OH stretching vibrations in EVA. Despite that the vinyl acetate monomer does not have -OH group in its structure, there is an interaction between the carbonyl group with a neighbor hydrogen atom for forming the hydrogen bonds and conjugated carbonyls, either with the methylene of the vinyl acetate or the ethylene monomer (Fig S.3.) (Ramírez-Hernández et al. 2019).
The XRD of PVC/EVA/CS/Str nanofibers is demonstrated in Fig S.4. In this pattern spectra, there is a broad peak at approximately 2θ = 24˚, that is representative of amorphous structure of PVC/EVA/CS/Str nanofibers. In other words, there is no crystalline phase for these nanofibers.
The evaluation of the porosity and special surface area of the PVC/EVA and the composite of PVC/EVA/CS/Str electrospun nanofibers were carried out by nitrogen sorption assessments at 77 K. The special surface area is enhanced from 9.2 m2 g− 1 to 39.8 m2 g− 1. Besides, the PVC/EVA/CS/St composite nanofiber illustrated a larger pore volume of 0.2754 cm3 g− 1 compared to PVC/EVA (0.021 cm3 g− 1).
Consequently, the prepared PVC/EVA/CS/Str electrospun nanofibers are suggested be an excellent candidate for ATZ and AMT adsorption.
3.2. Optimization of TF-µSPE conditions
To obtain the optimal conditions, the influence of principal alternatives (pH, adsorption time, NaCl concentration %(w/v), sorbent type, sorbent size and stirring rate, desorption time, eluent type and eluent volume) was evaluated by using one at time method and design-expert.
The response surface method (RSM) was used for optimization with design-expert 12.0 software because this method has some advantages: (1) with a small number of experimental units, it prepares maximum information, (2) possibility to utilize and change parameters simultaneously, (3) obtaining the interaction between the factors and removing not significant levels, (4) ability to transform when the residuals are not normally distributed and etc. Among of RSMs that were existed for optimization, such as, Box Behnken design (BBD) and central composite design (CCD), the CCD was selected. This design in compare with BBD has difference in number of levels. While five levels are utilized in CCD, three levels are utilized in BBD (KESKİN et al. 2021). Also, the total number of design point is:
N = 2k + 2K + C0 (K is number of factors and C0 is repetition number of center points) and
N = 2K(K-1) + C0 (K is number of factors and C0 is number of center points) for CCD and BBD, respectively. N in this design is = 23+ 2×3 + 5 = 19 (Zolgharnein et al 2013).
The equation of this design is: R(adsorption) = 0.9502 − 0.0110 A-0.0033 B + 0.0049 C + 0.0173 AB -0.0075 AC -0.0152 BC -0.0599 A2 + 0.0020 B2 -0.0623 C2. The optimum values of the pH of adsorption solution, time of adsorption, and percent of supporting NaCl concentration were obtained at 6.5, 15 min, and 15%(w/v), respectively (Sd = 0.0473 and R2 = 0.8923). The pKa values of ATZ and AMT are 1.68 and 4.0, respectively (Frías et al. 2004). Thus, is can be concluded that both ATZ and AMT have weak basic strength containing amino group while the EVA latex is copolymer that was obtained from polymerization of EVA monomers which have the carboxylic acid groups; so, it has a weak acidic strength. Under the media with acidic pH, the ATZ and AMT were in their protonated form, while at more basic pH values, the acidic group of EVA was in ionic form. Both of these conditions will weaken the hydrogen bond between the intended analytes and the adsorbent (Yin et al. 2022). 3D surface of the model graphs for effective factors on ATZ and AMT adsorption are shown in Fig. 3. A. Also, primary information for designing of CCD and tis statistics for the optimization of the effective factors on adsorption of ATZ and AMT on PVC/EVA/CS/Str electrospun nanofibers, were shown in Table S.2. and Table S.3., respectively.
Other effective factors on the adsorption process were optimized by applying one at time method (Fig S.5.A-C.). According to Fig S. 5. A. various sorbent size (1× 0.5, 1×1, 1×1.5, 1×2 and 1× 2.5 cm) was investigated. when the sorbent size was equal to 1×1 cm, the peak area was highest rather than those related to other sorbent sizes. Eventually, 1×1 cm sorbent size was selected as the optimum size. Also, Fig S.5. B. represents the effect of different nanofiber composites such as PVC/EVA, PVC/EVA/CS and PVC/EVA/CS/str. The optimum composite as can be observed, was PVC/EVA/CS/Str. Also, the PVC/EVA composite has lowest CD-IMS peak area as it does not have any hydroxyl groups while the plenty of mentioned functional groups exists in both of the CS and Str which can form hydrogen bonds with the target analytes (Moradi et al. 2019). CS also has aromatic ring and other functional groups such as amine (N-H) and carbonyl (C = O). Therefore, the CS and Str increase the sorption capacity significantly. In the following, sample agitation was investigated with applied vary of stirring rates (100, 150, 200, 250, 300, 350 and 400 rpm). The obtained results demonstrated that the adsorption process of ATZ and AMT on the PVC/EVA/CS/str electrospun nanofibers needed to neither very slow nor very fast rates, and this process was accomplished with the medium stirring rate. Hence, the stirring rate = 250 rpm was chosen as the optimum adsorption rate (Fig S.5.C.). To survey the effect of the type of eluent, the different eluents including ethanol, methanol, acetone and chloroform were used. As can be seen in Fig S.6.A. methanol as had the highest peak area of CD-IMS for both ATZ and AMT. In addition, to appraise the influence of the eluent volume, the volumes of 200, 300, 400, 500 and 600 µL were examined (Fig S.6.B.) Satisfactory extraction results for the desired analytes were obtained with 400µL of elution volume. Besides, desorption time was varied from 2.0 to 12.0 min and according to the results shown in Fig S.6. C., after 5.0 min, a considerable decrease in extraction efficiencies was observed.
3.3. Reusability of PVC/EVA/CS/str nanofibers
In order to evaluate the durability of the PVC/EVA/CS/Str electrospun nanofibers, the efficiencies of extraction of these nanofibers after 20, 30, 40, 50, 60, 70 and 80 times of utilizing nanofibers in standard solution were compared (Fig S.7.). As can be seen the electrospun nanofibers can be used 60 times without a considerable loss of frequency.
3.4. Validation of the proposed method
The merits figures of TF-µSPE method based on PVC/EVA/CS/Str electrospun nanofibers as a sorbent were evaluated. These figures are including of linear dynamic range (LDR), limit of detections (LODs), regression coefficients (R2), enrichment factors (EFs) and relative standard deviations (RSDs) of intra-and inter-day (Table 1.). The wide LDR in the range of 8.0-300.0 ng.mL− 1 with R2\(\ge\) 0.9821 was obtained for the analysis of the desired analytes. LODs values based on S/N = 3 were reached 2.5 and 2.3 ng.mL−1 for ATZ and AMT, respectively. The accuracy of current method was estimated with utilizing the spiked samples at four various concentration levels (10.0, 50.0, 100.0 and 250.0 ng.mL− 1). The RSD of intra- and inter day values were in ranging from 4.6–5.3% and 5.1–5.9%, respectively that represent the acceptable precision of this method. To compute the EF values in this study, firstly, the mixing of the ATZ and AMT standard solutions were injected to IMS, then after exposing the standard solution to the extraction process, the desorption solution was injected to IMS. After that, the two series of injection, the calibration curve was drawn to investigate the concentration effect, and also to obtain the relationship between concentrations and peak areas. The EFs were obtained from the slope ratio of the calibration curves (obtained slope of extraction calibration curve/obtained slope of direct calibration curve). These ratios were obtained 28.7 and 25.0 for ATZ and AMT, respectively.
Table 1
Analytical characteristics of the TF-µSPE-CD-IMS for the determination of ATZ and AMT under the optimum conditions.
Analyte
|
LODa
|
LDRb
|
Regression equation
|
R2
|
EFc
|
%RSD (Intra-day)d
|
%RSD (Inter-day)d
|
10a
|
50
|
100
|
250
|
10
|
50
|
100
|
250
|
ATZ
|
2.5
|
8.0-300.0
|
y = 0.376x + 0.00001
|
0.9821
|
28.7
|
4.6
|
4.9
|
5.0
|
5.2
|
5.1
|
5.3
|
5.6
|
5.9
|
AMT
|
2.3
|
8.0-300.0
|
y = 1.350x + 0.00005
|
0.9996
|
25.0
|
4.8
|
5.0
|
4.9
|
5.3
|
5.7
|
5.3
|
5.4
|
5.6
|
a Limit of detection |
b Linear dynamic range |
c Enrichment factor |
d Relative standard deviation |
3.5. Real sample analysis
To evaluate the capability of this method for the determination of trace amounts of ATZ and AMT, real samples such as lettuce, tomato and milk samples were used. First, sample preparation was done for each real samples according to section 2.3.4. All the real samples were kept in refrigerator at -4°C until analysis. The real samples were spiked with 10, 50, 100 and 250 ng.mL− 1 of both analytes. Figure 4. A. and Fig. 4.B. represent the ion mobility spectrums of the non-spiked and spiked lettuce and tomato samples, respectively. In addition, Table 2. exhibits the results of three replicate analyses of three samples of each real sample obtained with the proposed method. The relative standard deviations (RSD%) for the measurements of ATZ and AMT were ≤ 7.0%. Relative recoveries (RR%) are ranging from 88.6 to 98.1% for ATZ and AMT, respectively. RR% in three real samples were obtained according to the equation (Eq. 1):
%RR =\(\frac{{\text{C}}_{\text{f}\text{o}\text{u}\text{n}\text{d}}-{\text{C}}_{\text{r}\text{e}\text{a}\text{l}}}{{\text{C}}_{\text{a}\text{d}\text{d}\text{e}\text{d}}}\)× 100 (Eq. 1.)
Table 2
Analytical results for the extraction and determination of ATZ and AMT in real samples under the optimum conditions.
Sample
|
Analyte
|
C added (µg. L− 1)
|
C found (µg. L− 1)
|
%RSDa
|
%RRb
|
%MEc
|
Lettuce sample 1
|
ATZ
|
-
|
9.6
|
5.6
|
-
|
-
|
10.0
|
18.8
|
5.3
|
92.1
|
92.6
|
50.0
|
56.4
|
5.2
|
93.6
|
94.1
|
100.0
|
104.1
|
4.9
|
94.5
|
95.3
|
250.0
|
250.4
|
5.0
|
96.3
|
96.0
|
AMT
|
-
|
11.1
|
6.3
|
-
|
-
|
10.0
|
20.1
|
6.2
|
89.6
|
90.1
|
50.0
|
58.7
|
5.4
|
92.1
|
92.5
|
100.0
|
104.3
|
5.6
|
93.2
|
93.6
|
250.0
|
243.4
|
5.5
|
92.9
|
93.0
|
Lettuce sample 2
|
ATZ
|
-
|
8.8
|
5.9
|
-
|
-
|
10.0
|
18.0
|
5.8
|
91.6
|
92.1
|
50.0
|
55.2
|
5.3
|
92.8
|
93.4
|
100.0
|
102.6
|
5.1
|
93.8
|
95.1
|
250.0
|
244.6
|
5.6
|
94.3
|
94.6
|
AMT
|
-
|
11.9
|
5.7
|
-
|
-
|
10.0
|
21.0
|
6.1
|
91.2
|
91.6
|
50.0
|
57.8
|
6.2
|
91.8
|
92.0
|
100.0
|
105.5
|
5.8
|
93.6
|
94.3
|
250.0
|
245.4
|
5.4
|
93.4
|
93.9
|
Lettuce sample 3
|
ATZ
|
-
|
9.1
|
6.0
|
-
|
-
|
10.0
|
18.2
|
5.9
|
90.9
|
89.9
|
50.0
|
55.9
|
5.4
|
93.6
|
92.6
|
100.0
|
102.8
|
5.6
|
93.7
|
94.6
|
250.0
|
246.4
|
5.1
|
94.9
|
95.0
|
AMT
|
-
|
20.0
|
7.0
|
-
|
-
|
10.0
|
29.3
|
6.5
|
92.6
|
93.1
|
50.0
|
66.5
|
6.3
|
92.9
|
92.6
|
100.0
|
113.6
|
6.2
|
93.6
|
93.6
|
250.0
|
265.3
|
6.3
|
98.1
|
95.9
|
Tomato sample 1
|
ATZ
|
-
|
11.1
|
5.6
|
-
|
-
|
10.0
|
20.5
|
5.5
|
94.1
|
94.3
|
50.0
|
58.7
|
5.4
|
95.2
|
96.0
|
100.0
|
107.4
|
5.2
|
96.3
|
96.6
|
250.0
|
250.1
|
5.0
|
95.6
|
96.0
|
AMT
|
-
|
16.7
|
6.1
|
-
|
-
|
10.0
|
25.9
|
6.2
|
92.4
|
93.1
|
50.0
|
62.7
|
5.6
|
91.9
|
93.6
|
100.0
|
110.2
|
5.4
|
93.5
|
94.5
|
250.0
|
246.7
|
5.2
|
92.0
|
93.9
|
Tomato sample 2
|
ATZ
|
-
|
11.9
|
4.9
|
-
|
-
|
10.0
|
20.8
|
4.8
|
88.9
|
89.9
|
50.0
|
58.7
|
5.1
|
93.6
|
93.6
|
100.0
|
104.9
|
4.9
|
93.0
|
93.9
|
250.0
|
250.9
|
4.8
|
95.6
|
94.6
|
AMT
|
-
|
19.0
|
5.3
|
-
|
-
|
10.0
|
28.0
|
5.4
|
90.4
|
89.4
|
50.0
|
64.8
|
5.2
|
91.6
|
90.2
|
100.0
|
111.7
|
5.6
|
92.7
|
93.6
|
250.0
|
253.0
|
5.0
|
93.6
|
92.8
|
Tomato sample 3
|
ATZ
|
-
|
20.0
|
6.3
|
-
|
-
|
10.0
|
29.3
|
6.0
|
92.4
|
93.0
|
50.0
|
65.8
|
5.9
|
91.6
|
92.6
|
100.0
|
113.8
|
6.0
|
93.8
|
92.9
|
250.0
|
257.5
|
6.1
|
95.0
|
94.1
|
AMT
|
-
|
32.0
|
4.8
|
-
|
-
|
10.0
|
40.9
|
5.1
|
89.0
|
90.2
|
50.0
|
78.1
|
5.0
|
92.1
|
93.2
|
100.0
|
127.6
|
5.3
|
95.6
|
95.2
|
250.0
|
270.5
|
5.0
|
95.4
|
96.3
|
Milk sample 1
|
ATZ
|
-
|
8.5
|
6.2
|
-
|
-
|
10.0
|
17.7
|
6.3
|
92.4
|
91.6
|
50.0
|
54.9
|
6.0
|
92.9
|
92.7
|
100.0
|
102.5
|
5.8
|
94.0
|
90.9
|
250.0
|
247.5
|
6.0
|
95.6
|
93.6
|
AMT
|
-
|
17.9
|
5.8
|
-
|
-
|
10.0
|
26.8
|
6.1
|
88.6
|
90.2
|
50.0
|
62.7
|
5.6
|
89.6
|
93.6
|
100.0
|
109.3
|
5.5
|
91.4
|
94.6
|
250.0
|
247.9
|
5.2
|
92.0
|
94.8
|
Milk sample 2
|
ATZ
|
-
|
9.0
|
5.7
|
-
|
-
|
10.0
|
18.4
|
6.3
|
94.3
|
94.1
|
50.0
|
56.1
|
6.0
|
94.1
|
94.5
|
100.0
|
104.6
|
5.5
|
95.6
|
93.8
|
250.0
|
248.3
|
5.2
|
95.7
|
96.1
|
AMT
|
-
|
19.0
|
5.3
|
-
|
-
|
10.0
|
28.1
|
5.3
|
91.0
|
89.6
|
50.0
|
65.3
|
5.4
|
92.6
|
89.6
|
100.0
|
111.8
|
5.0
|
92.8
|
91.2
|
250.0
|
251.3
|
5.1
|
92.9
|
93.4
|
Milk sample 3
|
ATZ
|
-
|
8.8
|
6.4
|
-
|
-
|
10.0
|
17.8
|
6.2
|
90.1
|
91.3
|
50.0
|
54.9
|
6.3
|
92.1
|
92.8
|
100.0
|
101.7
|
5.9
|
92.9
|
94.6
|
250.0
|
241.6
|
5.8
|
93.1
|
95.3
|
AMT
|
-
|
18.4
|
6.5
|
-
|
-
|
10.0
|
27.5
|
6.2
|
90.8
|
90.4
|
50.0
|
64.9
|
5.6
|
93.1
|
91.0
|
100.0
|
111.3
|
5.5
|
92.9
|
92.9
|
250.0
|
254.2
|
5.4
|
94.3
|
92.8
|
a Relative standard deviation (n = 3) |
b Relative recovery |
c Matrix effect |
In which that, Cfound is concentration of the target analytes in the eluent after adding a specific amount of standard solution, Creal is the concentration of the analyte in the real samples before the TF-µSPE process and Cadded is the concentration of standard solution that was spiked to real sample at each spiked process.
3.6. Matrix effect evaluation
Since the interference species that exist in the complex matrix of the analytes, could interact with PVC/EVA/CS/Str electrospun nanofibers and influence on the capacity of adsorption of the nanofiber, the Matrix effect percent (%ME) was obtained for all repeated of three real samples under the optimized conditions utilizing the following equation:
%ME= \(\frac{{\text{A}}_{\text{s}}-{\text{A}}_{\text{n}}}{{\text{A}}_{\text{w}}}\) (Eq. 2.)
In which that As, An and Aw are obtained signal for spiked, non-spiked and blank sample, respectively. The values of %ME was represented in Table 2. These results show the analysis of ATZ and AMT in different real samples (lettuce, tomato and milk) is not affected with interference species significantly.
3.7. Comparison with other reported approaches
The comparison of the proposed approach with other approaches that have been reported in literature for analysis of ATZ and AMT are presented in Table 3. The current method has good repeatability and linearity. This comparative table also shows that the extraction time at current work with TF-µSPE is lower than other methods. Altogether, this procedure is simple, fast, sensitive and practical technique which can be used for measuring ATZ and AMT in samples with aqueous media.
Table 3
Comparison of the TF-µSPE-CD-IMS with other extraction techniques for determination of ATZ and AMT.
Analytes
|
Sample type
|
Technique
|
LOD (ng.mL− 1)
|
LDR (ng. mL− 1)
|
Extraction time (min)
|
Ref.
|
ATZ
AMT
|
water
water
|
IMS
|
0.013
0.015
|
0.05-10
0.05-10
|
60
|
Zou et al, (2016)
|
ATZ
AMT
|
water
water
|
LC-EIS-MS/MS
|
0.07
0.09
|
Up to 10
Up to 10
|
-----
|
Silva et al. (2013)
|
ATZ
AMT
|
water
water
|
IMS
|
15
10
|
50-2800
50-2800
|
45
|
Mohammadi et al. (2009)
|
ATZ
AMT
|
water
|
GC-MS
|
0.09
0.05
|
0.5–500
0.1–500
|
-----
|
Bagheri et al. (2016)
|
ATZ
AMT
|
Lettuce, milk, tomato
|
IMS
|
2.5
2.3
|
8-300
8-300
|
20
|
This work
|