Introduction

Sinapic acid and tyrosol (Fig. 1) are common constituents of plants and fruits. These substances can be found for example in cranberry, wine, mustard seeds, and selected types of oils [1,2,3]. Tyrosol is also one of the main natural phenols in argan oil [4]. As antioxidants, they can protect cells against oxidation [4, 5]. Even though they are not as potent as other antioxidants, their higher concentration and good bioavailability indicate that they may have an important overall effect in the antioxidant properties of natural products. This effect may contribute significantly to the health benefits for example of olive oil and, more generally, the Mediterranean diet [5].

Fig. 1
figure 1

Structure of sinapic acid (a) and tyrosol (b)

Full size image

Several methods, mainly based on cyclic voltammetry and differential pulse voltammetry, have been reported for the determination of sinapic acid [6] or tyrosol [7]. However, to the best of our knowledge, only a few analytical methods have been reported for the simultaneous determination of sinapic acid and tyrosol including HPLC with UV or MS detection [8,9,10].

Recent publications have demonstrated that flow-injection analysis (FIA) with multiple-pulse amperometric (MPA) detection could be used for simultaneous measurement of two or more electroactive species [11,12,13]. An aliquot of sample solution is directly injected into a FIA system and the compounds are selectively detected at a single working electrode by applying two sequential potential pulses. A simple correction factor must be used for the calculation. This approach has some important advantages: it is inexpensive and simple, and has small sample and reagent consumption (with reduction in waste generation) and high sampling rates [14]. This strategy was used for simultaneous amperometric detection of sugars [15], drugs [16,17,18], antioxidants [19], synthetic colorants [20], as well as for the use of internal standard method in FIA [21].

This paper demonstrates that MPA detection in combination with FIA on a glassy carbon electrode (GCE) can be used for the simultaneous determination of sinapic acid and tyrosol. Results obtained with this method were evaluated with respect to recovery, repeatability, linearity, and detection limits.

Results and discussion

The influence of pH on the cyclic voltammograms of oxidizable sinapic acid and tyrosol (both at 0.1 mM) was investigated in a mixed methanol and 0.040 M B-R buffer (1:9, v/v) medium. Both sinapic acid and tyrosol have in this medium single peak, whose position and height depend on the pH. The dependence of Ep on pH for sinapic acid can be described using linear regression as Ep (mV) = − 47.1 pH + 773.2 (r > 0.98). In the case of tyrosol, the dependence can be described as Ep (mV) = − 58.5 pH + 1098.2 (r > 0.99). In both cases, with rising pH, there is a rapid drop in the peak heights. The carrier solution of pH 2.0 was selected for further experiments, because in this medium, the oxidation peaks of sinapic acid and tyrosol were well separated (> 350 mV) and the peaks were highest in the CV experiments; addition of methanol was necessary due to the low solubility of these phenolic antioxidants in water. To identify the optimal oxidation potentials to perform simultaneous determination of sinapic acid and tyrosol, hydrodynamic voltammograms were first obtained separately for each compound using MPA-FIA (Fig. 2). In this case, standard solutions containing sinapic acid or tyrosol (0.1 mM) were injected into the system. Sequential potential pulses of 100 ms duration from + 0.40 to + 0.80 V for sinapic acid and from + 0.70 to + 1.10 V for tyrosol were applied continuously; the current at each potential pulse was monitored and used to construct the hydrodynamic voltammogram for the electrochemical oxidation of both compounds.

Fig. 2
figure 2

Hydrodynamic voltammograms obtained by plotting peak current values as a function of the corresponding applied potential pulses. The solutions contained sinapic acid (filled squares) or tyrosol (filled circles) (both at 0.1 mM). Carrier solution: methanol, 0.040 M B-R buffer pH 2.0 (1:9, v/v); injected volume: 100 mm3; flow rate: 1.0 cm3 min−1

Full size image

It can be seen that the peak potentials of the obtained curves differ enough to enable the selective determination of the analytes. Namely, potentials between + 0.70 and + 0.80 V would only cause the oxidation of sinapic acid without significant interference from tyrosol; therefore, + 0.75 V (100 ms) was selected as the first potential pulse. Potential of + 1.10 V (100 ms) was selected as the second potential pulse, where both target analytes are fully oxidized. Tyrosol can be quantified if the current from the oxidation of sinapic acid at + 1.10 V is previously subtracted. However, direct subtraction of the current response at + 0.75 V (exclusive oxidation of sinapic acid) from the current response at + 1.10 V (oxidation of both target analytes) is not possible, as the current responses detected for sinapic acid at + 0.75 and + 1.10 V are not equal, and a correction factor (CF) must be used. For tyrosol determination at + 1.10 V without interference from sinapic acid, the CF can be obtained by a simple injection of a standard solution containing only sinapic acid and by the application of the following equation:

$$CF = {I_{{\text{sinapic acid}} + 1.10\;{\text{V}}}}/{I_{{\text{sinapic acid}} + 0.75\;{\text{V}}}}.$$
(1)

Then, the current originating from tyrosol oxidation at + 1.10 V during the analysis of solution containing both sinapic acid and tyrosol can be calculated using the equation:

$${I_{\text{tyrosol}}} = {I_{ + 1.10\,{\text{V}}}} - (CF \times {I_{{\text{sinapic acid}} + 0.75\,{\text{V}}}}).$$
(2)

In the development of the proposed method, an additional parameter should be considered: the CF value must be constant in the selected concentration interval. In the concentration interval between 10 to 100 μM of sinapic acid, this requirement was fulfilled and the CF value was calculated as 1.10 ± 0.06 (n = 3).

Other FIA parameters were optimized to obtain the highest signal for sinapic acid and tyrosol. Figure 3 illustrates the dependence of the peak current of the analytes on the injected volume (Fig. 3a) and flow rate (Fig. 3b). The optimization of the injected volume was done similar to any FIA system, i.e., the optimal injection volume was selected as the maximum volume above which the peak height does not further increase. The effect of the injected sample volume (Fig. 3a) on the MPA response was investigated in the range from 20 to 150 mm3, using solutions of each antioxidant and applied electrode potentials of + 0.75 V for sinapic acid and + 1.10 V for tyrosol. The amperometric signal increased with the injected sample volume up to 100 mm3 and then remained almost constant for higher injected volumes; this value was thus selected for a subsequent measurements. The effect of the flow rate (Fig. 3b) was evaluated by varying its values from 1.0 to 5.0 cm3 min−1, using injected volume 100 mm3 and applying electrode potentials of + 0.75 V for sinapic acid and + 1.10 V for tyrosol. The electrode response increased with flow rate up to 3.0 cm3 min−1 and then remained almost constant for higher flow rates; thus, this value of flow rate was selected for further amperometric measurements, due to lower consumption of the carrier solution and higher peaks.

Fig. 3
figure 3

Optimization of FIA parameters: dependence of current response on a injected volume and b flow rate of sinapic acid (filled squares) or tyrosol (filled circles) (both 0.1 mM). Potential pulses: + 0.75 (sinapic acid) and + 1.10 V (tyrosol) of 100 ms duration; carrier solution: methanol − 0.040 M B-R buffer pH 2.0 (1:9, v/v)

Full size image

To examine the stability of the analytical signal, a repeatability study was conducted (Fig. 4); under the optimized conditions, ten successive injections of a standard solution containing sinapic acid and tyrosol (both 0.1 mM) were carried out. The results demonstrate that the MPA-FIA system provides good repeatability (RSD < 4.0%, n = 10) and a high sampling rate (around 170 determinations h−1). From the same figure, difference between the values of the baseline for each inserted potential may be observed. In addition, other publications mentioned earlier obtained similar results [12,13,14]. This problem has a great connection with the length of the individual pulses, the size of the inserted potential pulses, and the magnitude of the potential difference between the individual pulses, which are in very fast sequences during the measurement only the carrier solution or carrier solution and analytes [22, 23].

Fig. 4
figure 4

Repeatability data obtained from successive injections of a solution containing sinapic acid and tyrosol (both 0.1 mM) (n = 10). Potential pulses: + 0.75 (sinapic acid) and + 1.10 V (tyrosol) for 100 ms each; carrier solution: methanol − 0.040 M B-R buffer pH 2.0 (1:9, v/v); injected volume: 100 mm3; flow rate: 3.0 cm3 min−1

Full size image

Using the optimized experimental conditions selected for the determination of sinapic acid and tyrosol, analytical figures of merit were obtained using solutions containing varying concentrations of one antioxidant, while the concentration of the other antioxidant remained constant. Figure 5 illustrates the amperometric responses of this measurement at one concentration level (100–10 μM) and the calibration plots proving the proportionality between the amperometric current and the concentrations of the analytes. Linear regression of these two series of experiments leads to excellent correlation coefficients (r > 0.99, in both cases) and the LQ values obtained for these antioxidants are at micromolar level (see Table 1). The course of determination should be without major complications in the case of the measurement of sinapic acid and tyrosol in matrices mentioned earlier with a high proportion of these substances. In the case of the rest of real samples, complications associated with the presence of other antioxidants, which naturally occurring in the real matrices can arise. The interference of other antioxidants depends highly on their properties, namely ascorbic acid, and most other antioxidants oxidize earlier than the measured analytes using the given conditions. This would change the procedure in the next step, namely recalculation of the peak heights of the determined substances by the correlation factor as explained earlier. A minor disadvantage may be that, for each real sample, a specific method for determination of the mentioned analytes would have to be developed.

Fig. 5
figure 5

MPA-FIA recordings obtained after injections of a six standard solutions (100–10 μM) of tyrosol + sinapic acid (0.1 mM) and b six standard solutions (100–10 μM) of sinapic acid + tyrosol (0.1 mM). Inset shows calibration curves for tyrosol (filled circles) and sinapic acid (filled squares). For measurement conditions, see Fig. 4

Full size image
Table 1 Figures of merit of the proposed method for the simultaneous MPA-FIA determination of sinapic acid and tyrosol
Full size table

Conclusion

The present work demonstrates the possibility of simultaneous determination of sinapic acid and tyrosol using a flow-injection system with multiple-pulse amperometric detection. The advantages of the technique are short time of analysis (170 injections h−1), low consumption of samples and reagents, high precision (RSD < 4.0%; n = 10), and linear calibration curves (r > 0.99). The limits of quantification were 0.86 and 1.03 μM for sinapic acid and tyrosol, respectively. This method has a good potential to be applied in routine analysis in substitution of expensive chromatographic separation systems.

Experimental

Sinapic acid (CAS number 530-59-6) and tyrosol (CAS number 501-94-0) were supplied by Sigma-Aldrich. Their individual stock solutions (c = 1.00 mM) were prepared by dissolving the exact amount of the respective substance in methanol (Merck Millipore, Germany) and they were kept at 4 °C. More diluted solutions were prepared by exact dilution of the stock solutions with mixture of methanol and 0.040 M Britton-Robinson (B-R) buffer (1:9, v/v). All electrochemical measurements were carried out in the same solution. The B-R buffer was prepared by mixing 0.20 M sodium hydroxide (Lach-Ner Neratovice, Czech Republic) with acidic solution consisting of 0.040 M boric acid (Lach-Ner Neratovice, Czech Republic), 0.040 M phosphoric acid (Merck Millipore, Germany), and 0.040 M acetic acid (Merck Millipore, Germany). All chemicals used for buffer preparation were of analytical grade purity. Distilled water was provided from a Mega-Pure 3A Liter Automatic Distillation System, USA.

Instrumentation and apparatus

All electrochemical recordings were performed using an Autolab PGSTAT12 potentiostat/galvanostat, controlled by NOVA version 1.11.2 software (Metrohm, Switzerland) working under Windows 7 (Microsoft Corporation). The three-electrode wall-jet configuration included a glassy carbon working electrode (GCE) (Metrohm, Switzerland, diameter of 2 mm and geometric area 3.1 mm2), a platinum wire, 1 cm in length and 0.5 mm in diameter, as a counter electrode, and an Ag/AgCl (3 M KCl) electrode as a reference electrode (MonokrystalyTurnov, Czech Republic) [24]. Flow of the carrier solution was provided by peristaltic pump MINIPULS Evolution (Gilson, USA) and injection of the sample was performed with a six-way injection valve (VICI Valco Instruments, Canada) equipped with a 100 mm3 sample injection loop. An Orion 266S pH meter (Thermo Fisher Scientific, USA) equipped with a combined glass pH electrode was used for pH measurements. The pH meter was calibrated with aqueous standard buffer solutions at ambient temperature.

Procedures

Pre-treatment of the GCE was done by polishing with alumina powder suspension (0.1 μm) on a damp polishing cloth (Metrohm, Switzerland) before fixing to the flow cell. This procedure was performed at the beginning of the working day.

Hydrodynamic voltammograms of sinapic acid and tyrosol were obtained separately by application of nine sequential potential pulses (from + 0.40 to + 0.80 V for sinapic acid and from + 0.70 to + 1.10 V for tyrosol, pulse width 100 ms) in triplicate injections of standard solutions through the FIA system using the MPA technique. The same technique was used for simultaneous amperometric detection of sinapic acid and tyrosol, applying pulses + 0.75 V for 100 ms (sinapic acid) and + 1.10 V for 100 ms (tyrosol) continuously (total time of the potential waveform was 200 ms).

The peak height (Ip) was evaluated from the amperometric FIA recording. The limit of quantification (LQ) was calculated as the analyte concentration corresponding to a tenfold standard deviation of the respective response from ten consecutive determinations at the lowest measurable concentration [25].