Tin content in samples of canned fruits and vegetables was determined by hydride generation inductively coupled plasma atomic emission spectrometry (HG-ICP-OES), and it was compared with results obtained by standard method of flame atomic absorption spectrometry (AAS). Selected tin emission lines intensity was measured in prepared samples after addition of tartaric acid and followed by hydride generation with sodium borohydride solution. The most favorable line at 189.991 nm showed the best detection limit (1.9
A significant quantity of food and beverages arrive on the market in a robust form of tinplate packaging. Hermetically sealed can allow minimization of headspace oxygen and also keep a long and safe shelf life with minimal use of preservatives. However, the use of tinplate for food and beverage packing will result in some tin dissolving into food content. Tinplate corrosion depends on many factors including can material, nature of the can linings and coatings, and nature and acidity of the contacting foodstuff. Only limited reports are available on the toxicological effects of inorganic tin as present in canned foods, resultant from dissolution on the tin coating [
Determination of tin in canned food became very important because it gives information about the contamination process and provides help to increase canned food quality and safety. In order to evaluate tin concentration in canned food and beverages, several analytical methods are described in the recent literature: fluorimetric with use of surfactant reagents to increase the sensitivity [
Atomic spectrometry methods in analysis of tin level in canned foods presume decomposition of food samples which is usually performed by acid digestion or dry ashing methods following by measurements in flame or graphite furnace mode of AAS or ICP-OES. Although the preparation step could influence the complete recovery of Sn, the sensitivity of applied measurement technique is also essential in correct evaluation of tin content in foods. For example, flame atomic absorption is recommended by European Committee for Standardization (CEN) as a standard analytical method for the determination of tin in fruit and vegetables preserved in cans [
In the present work a method of the tin determination in canned food samples by HG-ICP-OES method is described. The aim of study was to select the most appropriate measurement conditions for tin determination, especially at low-concentration range where FAAS shows insufficient sensitivity. Commercially available samples of fruits and vegetables analyzed by both methods were preserved in variously manufactured tinplate, that is, interior surfaces of cans were fully or partially lacquered. Among the chemical and physical factors which influence tin transfer into canned food, a quality of tinplate protection is also essential. Therefore, HG-ICP-OES as sensitive method for low-content tin determination in canned fruit and vegetable samples should be also helpful in estimating the efficiency and resistance of tinplate package.
A
ICP-AES operating conditions.
Instrument | Prodigy high-dispersive ICP |
Spectrometer | High resolution echelle polychromator |
Large-format programmable array detector (L-PAD) | |
RF generator | 40 MHz “free running” |
Argon flow | Coolant: 18 L min−1 |
Auxiliary: 0.8 L min−1 | |
Nebulizer: 36 psi | |
Nebulizer | Pneumatic (glass concentric) |
Spray chamber | Glass cyclonic |
Hydride generator | Leeman Labs. Inc. Part No. 130-1070 |
Three channel peristaltic pump (0.9 mL min−1) | |
T connector | |
Reaction coil | |
Output power (1.3 kW) | |
Plasma viewing (Radial) | |
Replicates for each analysis run | 3 |
Sample uptake delay | 50 s |
Integration time | 40 s |
Emission lines | Sn (II) 189.991 nm |
Sn (I) 224.605 nm | |
Sn (I) 235.484 nm | |
Sn (I) 283.999 nm |
A
For the homogenization of food samples a
High-purity-deionised water (Milli-Q Element system, Millipore, USA) was used for the preparation of standard solutions and dilution of samples. In the sample digestion procedures, a hydrochloric acid of suprapure grade (30% m/v) was used. Single-element standard solutions of Sn 1000 mg L−1 (Plasma Pure, Leeman Labs, Hudson, NH, USA) was used for the preparation of calibration standard solutions and control of plasma line positioning.
All calibration standards for AAS measurement were prepared during instrument run by automated dilution of reference tin solution of 50 mg L−1 in 10% v/v HCl.
For the determination of tin by HG-ICP-OES, a fresh solution of NaBH4 0.8% m/v in NaOH 0.5% m/v was prepared. Tartaric acid for the reduction of tin was prepared by dissolution of 10 g of solid compound in 1 L of ultrapure water. Calibration solutions of tin were prepared in the range of 0.1–50.0 mg L−1 by dilution with 1% m/v tartaric acid solution to appropriate volume. Calibration blank contained only aqueous solution of tartaric acid. Method of standard addition (MSA) included aliquots of prepared samples in which a standard solution of tin was added. MSA solutions were diluted with tartaric acid solution. The final concentration range in MSA sample solutions was 0.1–50.0 mg L−1 of tin.
The twenty five samples of canned fruits and vegetables stored in original package at room temperature were opened, and the whole content was transferred into a
Atomic absorption measurements were performed in accordance with recommended standard procedure for the tin determination described in CEN/TS 15506:2007 [
For the purpose of HG-ICP-OES measurements, the calibration standards and 10-fold diluted samples were mixed with a solution of NaBH4 by three-channel peristaltic pump at a rate of 0.9 mL min−1. Hydride generator assembly scheme is shown in Figure
Hydride generator (HG) schematic.
Time resolved analysis (TRA) of Sn solution (1 mg L−1 in tartaric acid) by HG-ICP-OES.
Intensity measurements were performed in triplicate. The precision of signal measurements expressed as relative standard deviation was established in the range 0.1%–4.0%. Calibration curves for each Sn line were shown in Figure
Method detection limits (
Wavelength/nm | LOQ/ | |
---|---|---|
Sn (II) 189.991 | 1.9 | 6.4 |
Sn (I) 224.605 | 1.4 | 4.5 |
Sn (I) 235.484 | 6.3 | 20.0 |
Sn (I) 283.999 | 6.5 | 20.7 |
Calibration curves (HG-ICP-OES) in conventional mode.
Calibration curves (HG-ICP-OES) in method of standard addition mode (MSA).
HG-ICP-OES measurements of tin content in samples were performed at all selected emission lines. The precision of measurement based on RSD calculations from three replicates showed the smallest disturbance of signal at 235.484 nm (RSD 0.1%) and 189.991 nm (RSD 0.4%), compared to 283.999 nm (RSD 2.6%) and 224.605 nm (RSD 4.0%). The sensitivity of measurements should be examined from the obtained calibration curves and TRA signals. By aspirating the solution of low tin concentration, that is, 1 mg L−1 the better sensitivity of signal was noted at 283.999 nm and 189.991 nm lines (Figure
Tin content measured by flame AAS and HG-ICP-OES in different samples of canned fruits and vegetables is shown in Table
Tin content in canned fruit determined by AAS and HG-ICP-OES (mean of two replicate samples,
No. | Sample | Tinplate protection | AAS | HG-ICP-OES |
---|---|---|---|---|
(1) | Peach compote, less sweat | Lacquered | 68.9 ± 1.2 | 69.6 ± 1.1 |
(2) | Tomato puree, double concentrated | Lacquered | <5 | 2.5 ± 0.1 |
(3) | White beans, sterilized | Lacquered | <5 | 5.8 ± 0.1 |
(4) | Red beans, sterilized | Lacquered | <5 | 3.5 ± 0.1 |
(5) | Champignons | Lacquered | <5 | 0.44 ± 0.02 |
(6) | Champignons, sterilized | Lacquered | 77.2 ± 0.8 | 77.2 ± 0.7 |
(7) | Peeled plum tomatoes, sterilized | White lacquer | <5 | 10.9 ± 0.2 |
(8) | Fruit salad | Yellow lacquer | <5 | 1.21 ± 0.04 |
(9) | Pineapples compote | Tin side and bottom, lacquered lid and seam | 25.6 ± 0.1 | 28.8 ± 0.1 |
(10) | Pineapples compote | Tin side and bottom, lacquered lid and seam | 37.6 ± 0.2 | 40.5 ± 0.2 |
(11) | Pineapples compote | Tin side and bottom, lacquered lid and seam | 199.2 ± 3.2 | 199.2 ± 3.2 |
(12) | Pineapples compote | Tin side and bottom, lacquered lid and seam | 37.7 ± 3.7 | 41.2 ± 3.7 |
(13) | Pineapples compote | Tin side and bottom, lacquered lid and seam | 74.5 ± 1.1 | 75.5 ± 1.3 |
(14) | Pineapples compote | Tin side and bottom, lacquered lid and seam | 23.3 ± 1.1 | 25.7 ± 1.1 |
(15) | Pineapples compote | Tin side and bottom, lacquered lid and seam | 47.8 ± 0.5 | 47.8 ± 0.7 |
(16) | Pineapples compote | Tin side and bottom, lacquered lid and seam | 40.3 ± 0.7 | 40.4 ± 0.7 |
(17) | Peach halves peeled, compote | Tin side and bottom, lacquered lid and seam | 28.6 ± 0.5 | 29.6 ± 0.4 |
(18) | Peach halves in syrup, less sweat compote | Tin side and bottom, lacquered lid and seam | 82.8 ± 1.4 | 104.9 ± 1.8 |
(19) | Compote of mixed fruits | Tin side and bottom, lacquered lid and seam | 53.5 ± 0.9 | 54.82 ± 0.8 |
(20) | Pineapples compote | Tin side and bottom, lacquered lid and seam | 37.7 ± 0.7 | 41.2 ± 0.7 |
(21) | Apricot compote, less sweat | Tin side and bottom, lacquered lid and seam | 115.9 ± 1.4 | 115.6 ± 1.3 |
(22) | Apricot compote | Tin side and bottom, lacquered lid and seam | 81.0 ± 1.1 | 76.7 ± 1.4 |
(23) | Fruit cocktail in light syrup | Tin side and bottom, lacquered lid and seam | 117.3 ± 1.3 | 115.8 ± 1.4 |
(24) | Fruit cocktail in light syrup | Tin side and bottom, lacquered lid and seam | 108.6 ± 1.2 | 107.3 ± 1.3 |
(25) | Mandarin oranges whole segments | Tin side and bottom, lacquered lid and seam | 105.5 ± 1.3 | 104.8 ± 1.2 |
Generally, the results of flame AAS and HG-ICP-OES measurements showed that tin content in all samples did not exceed maximum permissive level of 200 mg kg−1 in foodstuffs. Only one sample (no. 11) of pineapples compote reached this level. Comparing the results obtained from two applied techniques, it should be noticed that for the most of samples measured by AAS and where tin content was above 5 mg kg−1, the similar results by HG-ICP-OES were obtained. The exceptions were samples of pineapples compote (no. 9, 10, 12, 14, and 20) and one sample of peeled plum tomatoes (No 7) with slightly higher tin concentration obtained by HG-ICP-OES. By knowing that samples for HG-ICP-OES determination include a tartaric acid which also prevents a hydrolysis of possibly present Sn (IV) species, observed difference is quite understandable. A sample of peach halves compote (no. 18) showed slightly higher Sn concentration when measured by HG-ICP-OES compared to concentration obtained by AAS. Since that longer delay from sample preparation to start of AAS measurement could lead to hydrolysis of analyte and forming of insoluble Sn (IV) compounds, a significant loss of Sn absorption signal might occur. HG-ICP-OES measurements include the use of highly reactive reducing agents which convert the majority of tin species in Sn (II) and favour the stannane gas formation. Therefore, the much more reliable value in such case is denoted to HG-ICP-OES measurement.
Tin content in samples from complete lacquered cans (nos. 2–5, 7, 8) measured by AAS shows limitation of method capability at low-concentration range (Table
The results obtained for the food samples from fully protected inner walls of cans are very useful in predicting a quality of protection. It is shown that tenth times less portions of Sn are present in protected cans than in non-protected or partially protected tinplate. The exceptions, measured as approximately 70 mg kg−1 of tin, are established in Samples 1 and 6. The studies concerning the quality of tinplate by SEM and EDS suggest that defects of lacquer result in tin exposure [
In a summary, analysis of tin content in canned fruits and vegetables was performed by use of flame AAS which is recommended as a standard analytical method and also by HG-ICP-OES. Analytical efficiency of HG-ICP-OES method was tested on several Sn emission lines. Low-detection limit, (1.9
The investigation was performed as a part of project 119-1191342-1083 financed by Republic Croatia Ministry of Science, Education, and Sport.