Development of SPE clean-up procedure for acrylamide determination from potato-based products by GC-MS/MS

2.5.1 Gas chromatography

The carrier gas was helium (constant flow rate at 1.6 mL/min). The sample injection volume was 1 µL, which was injected into a PTV injector, in a split mode at a split ratio of 1:10. The injector temperature was set to 220°C. The oven temperature was set to 65°C and maintained for 1 min, then increased on ramp 1 with 15°C/min up to 170°C, on ramp 2 with 5°C/min up to 200°C, and finally on ramp 3 with 40°C/min up to 240°C, maintaining at this temperature for 15 min (total run time is 30 min).

2.5.2 Mass spectrometry

The analyses were performed in the electron impact ionization operation mode (EI⁺ mode – 70 eV), with the acquisition mode “selected reaction monitoring – SRM” and ion scanning mode “product”. AA and the IS were identified by the ion fragments corresponding to the derivatized ions, 2-BPA (2-bromopropenamide) and 2-BP(¹³C₃)A. The precursor ions with m/z 151 and 154 fragmentation were achieved with argon (1 mTorr), leading to the formation of product ions (daughter) with m/z 70 (2-BPA) and 73 (2-BP(¹³C₃)A), being used for quantification. The AA concentration calculation in the test samples was based on the ratio of the peak area corresponding to the product ions, with m/z 70 and 73 for 2-BPA and 2-BP(¹²C₃)A, respectively.

2.6.1 The FC procedure

The method previously developed in the laboratory (FC) involves liquid–liquid extraction and purification with Carrez solutions and on a chromatographic column (300 mm × 11 mm I.D.) containing calcinated sodium sulfate and activated florisil. The first extraction step was performed with 30 mL hot water (60°C), added to 1.5 g sample (ratio sample:water = 1:20), followed by a purification step with Carrez I and II solutions (440 µL each) for protein and carbohydrate removal, and then by a centrifugation step (30 min at 6,000 × g at 5°C) in a centrifuge with cooling (5804R, Eppendorf, Germany). The obtained supernatant was derivatized with 7.5 g KBr, 20–100 µL HBr, and 10 mL saturated bromine–water solution (around 1.6%) and let on a shaking water bath below 4°C, for at least 2 h. After the end of the derivatization reaction, bromine in excess was removed by adding around 1,000 mL of 1 mol/L sodium thiosulfate, until the yellow color disappeared. The derivatized compound (2,3-DBPA) was extracted with 70 mL mixture of ethyl acetate:hexane (4:1, v/v). The AA derivatized extract was concentrated in two steps. In the first step, a vacuum evaporation system was used and the extract was concentrated to ∼2 mL. In the second step, the extract was evaporated to dryness under a nitrogen stream. The residue dissolved in 50 mL hexane was purified on a chromatographic column filled with activated florisil and calcinated sodium sulfate, previously conditioned with 20 mL hexane. The 2,3-DBPA derivative was eluted with 150 mL acetone, then concentrated until dryness and the residue was redissolved in 400 µL ethyl acetate and 40 µL triethylamine. The final extract (2-BPA) was filtered through a 0.2 µm regenerated cellulose microfilter (17 mm diameter, Spartan 13RC, Whatman GmbH, Dassel, Germany) directly in a vial and analyzed by GC-MS/MS in SRM mode.

The validation parameters for the FC procedure used for AA quantification from bread, biscuits, and other bakery products were described by Negoiță and Culețu (2016), and the method was characterized by good sensitivity, with a limit of detection (LOD) of 1.23 µg/kg (0.008 mg/L) and a limit of quantification (LOQ) of 3.7 µg/kg (0.025 mg/L), respectively. The relative standard deviations (RSD) for repeatability and reproducibility were 0.4–4.9% and 2.23–5.10%, respectively. The recovery was in the range of 97%–105%.

2.6.2 The SPE procedure

The SPE clean-up procedure involves the simultaneous achievement of these stages on two SPE cartridges. Experimental variants (Table 1) were carried out to establish the optimum conditions for extraction and purification by using SPE cartridges in order to obtain a maximum efficiency of the analyte extraction from the food matrix and to obtain a sufficiently cleaned-up extract for chromatographic analysis. Thus, samples were weighed (0.5 g) in centrifuge vials of 50 mL and were spiked with 440 µL IS (concentration of 1 mg/L), while water at 60°C and at room temperature (ratio 1:20) and hexane (5–10 mL) were added and samples were homogenized for 60 min in a vortex mixer. Next, the samples were centrifuged for 20 min at 6,000 × g at 5°C. In some variants, the aqueous phase separated was treated with Carrez solutions (I and II, 150 µL each) and after that, it was centrifugated. The supernatant (10 mL) obtained after centrifugation was passed through the SPE cartridges. Conditioning and washing of SPE 1 and SPE 2 cartridges were carried out, according to SR EN 16618:2015 (28). In order to establish the optimum elution conditions for AA extraction, different volumes (1–10 mL) of methanol:water mixture (60:40, v/v) were used. Elution and residual solvents were collected and the extract obtained was derivatized with 1 g KBr, 20 µL HBr, and 1 mL saturated bromine–water solution (around 1.6%). Derivatization took place on a shaking water bath below 4°C, for at least 1 h and at the end of the derivatization reaction, bromine in excess was removed by adding around 100 mL of 1 mol/L sodium thiosulfate. After derivatization, the extract was concentrated until dryness and the residue was redissolved in 400 µL ethyl acetate and 40 µL triethylamine. The final extract (2-BPA) was filtered through a 0.2 µm regenerated cellulose microfilter directly in a vial and analyzed by GC-MS/MS in SRM mode. All experiments were reported as the mean of minimum three replicate measurements.

A comparison of the two sample preparation procedures is presented in Figure 1.

Figure 1

Diagram of the sample preparation steps based on the FC procedure and optimal conditions for SPE procedure.

3.1.1 Influence of solvent volume used for AA elution

In order to determine the optimum volume for achieving a maximum elution of the AA extraction from the food matrix, the experimental variants presented in Table 1 (V1–V6) were realized. The extracts obtained by passing the solvent (in one- or two phases) through the SPE-2 cartridges (Isolute ENV⁺) were collected separately in vials and subjected to derivatization, concentration, and injection at GC.

From the results obtained (Table 2), it is observed that the maximum efficiency of AA extraction was achieved when the elution was carried out with a volume of 5 mL solvent. In V2, where a two-phase elution was carried out (5 mL + 5 mL), it was observed that after the first elution volume, the signal/noise ratio (S/N) and the AA peak area (A_AA) were the largest, and after the second elution volume addition, S/N and A_AA were smaller than those of the procedural blank realized with the same procedure. This denotes that AA was completely eluted with the first volume of 5 mL and the volume added in the second phase is in excess and inefficient. Thus, the volume of 10 mL of solvent in a single elution phase (V1) is too much for AA elution, as with this volume other compounds from the sorbent of SPE-2 cartridge are eluted, which give interferences and reduce S/N by about three times and A_AA by about 5–6 times, compared to the first elution volume of 5 mL in V2. The volume of 5 mL for the first elution phase in V2 is confirmed by the results of the tests performed under V3 conditions. In V3 variant, AA elution was performed with a 5 mL volume in a single phase, sufficient for AA elution, the highest values of the A_AA and S/N being obtained.

In order to acquire higher elution levels of AA, compared to the use of a volume of 5 mL solvent, in a single-phase elution, smaller volumes of elution (≤5 mL) (V4, V5, and V6) were also tested. Thus, in V4, the AA elution with a volume of 5 mL, in two elution phases determined for the first elution a S/N (36.907) close to that obtained when using 5 mL solvent (first elution V2 and elution from V3).

A small S/N (1.412) and A_AA (29.201) were determined for V4 when 1 mL was used for the second elution. In V5 and V6, it is observed that AA is eluted both in the first and in the second elution volumes. Thus, the volume of 5 mL of elution solvent used in a single phase is optimal for an efficient AA extraction through the SPE-2 cartridge.

3.1.2 Influence of water temperature on AA extraction

For AA extraction from the RM by applying the SPE procedure, tests were performed, where the water temperature used for extraction was: 60°C (V3) and room temperature (V7) was the variable factor (Table 1). AA is highly soluble in water, minimizing the dissolution of hydrophobic compounds in food, as well as co-extraction with other undesirable compounds, such as salts, proteins, and carbohydrates, which could interfere with AA, if not removed by purification. The results obtained (Table 3) for the tests carried out in the two variants were satisfactory, fulfilling the required conditions imposed for (a) and (b).

Since there are no significant differences between the C_V determined under V3 and V7 conditions and the C_RM value (Δ_m ≤ U_Δ) and that the difference between the values determined in the laboratory and the RM value was not greater than ±10%, the AA extraction can be applied under the two variants.

Considering that in V7 the number of additional steps is reduced (water heating and temperature measurement) compared to V3, for AA extraction the water extraction variant at room temperature (V7) will be selected.

Similar results were obtained by Saraji and Javadian (2019) who studied different temperatures for AA extraction ranging from 15 to 50°C and showed that the optimum extraction efficiency was obtained when the extraction took place at room temperature (25°C).

Some authors reported that AA extraction from cereal-based products was realized with water at room temperature (Pittet et al. 2004), while others (Dunovská et al. 2004) recommended hot water (60–80°C) for AA extraction from potato chips, as AA is stable at this temperature and the extraction efficiency can increase.

3.1.3 The volume of hexane used for defatting the sample

AA is practically insoluble in non-polar organic solvents such as hexane, heptane, and carbon tetrachloride. Because potato-based products uptake oil when fried, the obtained emulsion can influence the purification procedure. As a consequence, it can result in an overlapped peak with the analyte of interest, so defatting the matrices in parallel with the extraction step was performed. These experiments were realized with different volumes of hexane: 10 mL (V7) and 5 mL (V8).

The results obtained after comparing the mean concentration values determined under the conditions of variants V7 and V8 with the RM concentration value showed that the required conditions are accomplished, with no significant differences between the values (Linsinger 2010). It was observed that when using a volume of 10 mL hexane, the AA extracted in water was less colored, the cleaned-up extract on SPE cartridges was suitable compared to V8, vacuum not being necessary in the extraction procedure. The fat was better removed when a higher amount of hexane was used (V7), thus reducing the purification time of the extract by the SPE procedure. For these reasons, the conditions of the V7 variant will be considered optimal for AA extraction. Cheng et al. (2019) studied the fat extraction efficiency of hexane, cyclohexane, and toluene from fried samples and from all these solvents, hexane was chosen as the best defatting solvent. Also, Bermudo et al. (2006) as well as Fernandes and Soares (2007) stated that the fat removal with n-hexane from different food products, such as cereal-based products, French fries, and potato chips, ensures a more adequate contact with the water incorporated into the food matrix, finally improving the extraction efficiency of AA.

3.1.4 Influence of chemical purification of extracts with Carrez solutions

In order to achieve a clear extract, the chemical purification of aqueous phase of AA collected after extraction, homogenization, and centrifugation was treated with Carrez solutions (V7). The V9 without Carrez was also realized, as shown in Table 1. By adding potassium hexacyanoferrate trihydrate solution, K₄[Fe(CN)₆]·3H₂O (Carrez I), followed by a heptahydrate zinc sulfate solution, ZnSO₄·7H₂O (Carrez II), the proteins can be removed to the extent of 85% (Sun et al. 2003; Soares 2015). For both variants performed under the working conditions mentioned, the required conditions were accomplished, concluding that between the C_V determined and the C_RM there are no significant differences. The difference between the two variants is the fact that the experiment without Carrez solutions (V9) is less expensive, requiring a shorter preparation time, the need for less reagents, and therefore further experiments will be performed with the V9 conditions. Addition of Carrez solutions was also reported by Ciesarová et al. (2004), Gökmen et al. (2005), and Şenyuva and Gökmen (2005) for AA determination in potato chips and crisps and other cereal products.

A maximum efficiency of AA extraction from the matrix and a sufficiently cleaned-up extract were obtained when the extraction and purification of the AA extracts from the potato products were performed with water at room temperature, simultaneously with 10 mL hexane, without Carrez solutions, and the clean-up was realized on two SPE cartridges, using an optimum elution solvent volume of 5 mL (V9).

3.1.5 Evaluation of some validation parameters of the SPE procedure

Under the optimized extraction condition of SPE procedure, some validation parameters were evaluated. In terms of linearity, the calibration curves were linear over the two concentration ranges of 50–500 µg/L and 400–5,000 µg/L, respectively, with the correlation coefficient R² > 0.9992. The RSD for repeatability ranged between 2.05 and 4.21%. The interlaboratory reproducibility was evaluated by the z-score of −0.8, obtained by participation in the PT 3095 on French fries test material launched by the FAPAS program. The sensitivity of the method was evaluated by LOD and LOQ. A LOD of 6.94 µg/kg and a LOQ of 20.83 µg/kg, respectively, were obtained for the SPE procedure, fulfilling the required conditions set by the European Commission (2017). The recovery for the spiked samples ranged between 85.64 and 104.51%. The analysis of results showed that the newly developed method is characterized by both high precision and accuracy.

3.2.1 Comparative study of the SPE and the FC procedures

AA concentration of potato-based samples (chips and French fries) was determined, based on both procedures described, FC procedure (Figure 2) and by applying the optimum working conditions of the SPE clean-up procedure (Figure 3). The obtained results are presented in Table 4.

Figure 2

Chromatogram of chips (a) and French fries (b) samples obtained with the FC procedure. AA – acrylamide peak, IS – labeled AA peak.

Figure 3

Chromatogram of chips (a) and French fries (b) samples obtained with the SPE procedure developed. AA – acrylamide peak, IS – labeled AA peak.

As it can be seen in Table 4, for all analyzed samples, the extended uncertainty (U_Δ) is higher than the absolute difference (Δ_m), which means that there is no significant difference between the AA content obtained by using the FC and SPE procedures.

The procedures were compared by using a regression line and the results are shown in the graph from Figure 4, where on x axis are presented the results obtained by using the FC procedure, while on y axis are presented the results obtained by using the SPE procedure. By analyzing the AA content determined by using these procedures, it can be concluded that there is a close similarity between results (R² = 0.9999).

Figure 4

Graphical comparison of the two procedures used to extract AA from chips and French fries samples

Considering that there are no significant differences (Δ_m ≤ U_Δ) between results obtained by using the SPE and FC procedures, the new procedure developed can be used with confidence for AA determination from potato-based products.

Application of SPE cartridges in sample preparation allows the simultaneous extraction and clean-up of analyte, reduction of sorbent quantities (31%), solvent volumes (21%), amount of reagents (13% KBr; 10% bromine water), and the preparation time (50%), compared to the previous method of extraction and purification (FC), which is time consuming, and laborious involving several stages of manual handling (calcination, florisil weighing and activation, preparation and washing of chromatographic columns for purification, etc.), thus resulting in a cost reduction with ∼30%.

3.2.2 PTs

The functionality and accuracy of the results of the new SPE procedure were demonstrated by participation in 2019 in two PTs launched by the FAPAS program: FAPAS 3094 (AA in biscuit-cookie) with the FC procedure and FAPAS 3095 (AA in French fries –precooked) with the SPE procedure. The RMs received were processed by both procedures, yielding z-scores less than ±2 (Table 5).

The obtained results demonstrate the accuracy of the SPE procedure developed.