The Science of the Total Environment 257 Ž2000.

191 ᎐ 197

Aluminium levels in spices and aromatic herbs
Francisco F. Lopez, Carmen CabreraU , M. Luisa Lorenzo, ´ M. Carmen Lopez ´
Department of Nutrition and Bromatology, School of Pharmacy, Uni¨ ersity of Granada, Campus de Cartuja, E-18071 Granada, Spain

Abstract We evaluated the levels of aluminium in a total of 72 samples of 17 different spices and aromatic herbs that are widely consumed in Spain, and in the Mediterranean diet, in general. Aluminium was determined in the samples mineralized with HNO3 and V2 O5 , using electrothermal atomization atomic absorption spectroscopy as the analytical technique. The accuracy and precision of the proposed method was verified against an NBS-certified reference material. Precision, expressed as relative standard deviation, ranged from 1.10 to 4.07%. The results obtained from recovery studies were of 97.90 " 1.20. Aluminium concentrations ranged from 3.74 to 56.50 ␮ grg Ždry wt... The presence of this metal was detected in all the samples we analysed. ᮊ 2000 Elsevier Science B.V. All rights reserved.
Keywords: Aluminium; Spices; Aromatic herbs; Dietary intake

1. Introduction During recent years, interest concerning aluminium has considerably increased due to the knowledge about potential toxic effects of this element. Insolubility of some Al compounds and their bioavailability are questioned, as well as a possible relation between high Al content in tissues and neurodegenerative disorders, such as Alzheimer’s disease or other encephalopathies and some cases of osteomalacia ŽSlanina et al., 1986; Jagannatha and Valeswara, 1995; Liu and Nordberg, 1995; Storey and Masters, 1995.. Aluminium is widespread throughout nature,

Corresponding author. Tel.: q34-958-243863; fax: q34958-243869. E-mail address: ŽC. Cabrera..

air, water, plants, and consequently in all the food chain ŽBertholf et al., 1988; Gerhardsson et al., 1994.. Recent studies have shown that acid rain decreases the pH in soil, permitting Al mobility through groundwater by increasing its bioavailability ŽGerhardsson et al., 1994. which increased the interest of many researches in dietary intake ŽMinoia et al., 1994; Arruda et al., 1995a; Storey and Masters, 1995.. Studies carried out on human exposure to Al through the diet and the environment in a wide variety of geographic locations have shown values for the daily Al average intake ranging from 1 to 100 mg. The daily Al dietary average intake is reported to be 10 ᎐ 12 mg ŽBertholf et al., 1988.. Recent studies indicate that the relative bioavailability of this element varies extensively according to dietary intake ŽArruda et al., 1995a,b..

0048-9697r00r$ - see front matter ᮊ 2000 Elsevier Science B.V. All rights reserved. PII: S 0 0 4 8 - 9 6 9 7 Ž 0 0 . 0 0 5 2 0 - 9


F.F. Lopez ´ et al. r The Science of the Total En¨ ironment 257 (2000) 191 ᎐ 197

In Spain, aluminium has been determined in several foods and beverages ŽArruda et al., 1993, 1995a,b; Vinas et al., 1998, ˜ et al., 1995; Lopez ´ 2000., but additional data on Al content in other products such as spices and aromatic herbs widely consumed in the habitual diet, are required. Certain types of these aromatic herbs show a high capacity to accumulate heavy metals ŽGreger, 1985; Wang et al., 1991; Wilplinger et al., 1995; Castro, 1998; Kanias et al., 1998; Muller et al., 1998.. The Al determination presents numerous problems, such as sensitivity and selectivity of the method and additional contamination risk during the analytical process ŽManning et al., 1982; Berden ´ et al., 1994; Pomeranz and Meloan, 1994.. Because of its specificity, sensitivity, accuracy and precision, electrothermal atomization atomic absorption spectroscopy ŽETA-AAS., is the most frequently recommended analytical technique for Al determination in mineralized samples of food and beverages ŽManning et al., 1982; Eschnauer and Scollary, 1995; Taylor et al., 1997.. The objective of the present study is to establish the levels of aluminium in spices and aromatic herbs that are widely and habitually used in the Mediterranean diet. We optimized an analytical method for Al determination by ETAAAS previous mineralization of the samples with HNO3 and V2 O5. The accuracy and precision of the method was previously carried out with the aim of checking the suitability of the analytical technique for the determination of aluminium.

5000 chart recorder ŽBausch and Lomb, Rochester, NY, USA.. Graphite furnace ᎏ a Perkin-Elmer HGA-700 graphite furnace and pyrolytically-coated graphite tubes with a L’Vov platform were used. Analyses were performed in peak area mode to determine absorbance values. Argon of 99.99% purity with a 300-mlrmin flow was used as the internal gas. Acid digestion block ᎏ multiplace digestion block Selecta ŽSelecta, Barcelona, Spain..

2.2. Reagents The following reagents were used:

2. Material and methods 2.1. Apparatus

Nitric acid ᎏ Suprapure, 65% ŽE. Merck, Darmstadt, Germany.. Vanadium pentoxide ᎏ reagent grade ŽE. Merck.. Magnesium nitrate ᎏ reagent grade ŽE. Merck.; an aqueous solution of 0.5% Žwrv. was prepared. Ammonium molybdate ᎏ reagent grade ŽE. Merck.; a saturated solution was prepared. Aluminium standard solution ᎏ Titrisol solution Ž1.000 " 0.002 g. ŽE. Merck.; working solutions ranges of 10 ᎐ 50 ngrml. Water ᎏ all solutions were prepared with bidistilled deionized water with a specific resistivity of 18 M ⍀ и cm which was obtained by filtering twice distilled water through a Millipore Milli-Q purifier system model RO15 ŽMillipore, Gif-sur-Yvette, France. immediately before use.

2.3. Material The following apparatus were used:

Spectrophotometer ᎏ a Perkin-Elmer 1100B double beam atomic absorption spectrophotometer equipped with a deuterium background corrector was used with a slit width of 0.7 nm, with a 25-mA hollow cathode lamp ŽPerkin-Elmer, Norwalk, CT, USA.; and D-

To eliminate possible contamination from detergents or other sources, all glassware and polyethylene material was treated with a 30% Žvrv. HNO3 solution for 24 h and then rinsed several times with bidistilled deionized water immediately before use. An exhaustive cleaning of glassware material ŽManning et al., 1982; Berden ´ et al.,

F.F. Lopez ´ et al. r The Science of the Total En¨ ironment 257 (2000) 191 ᎐ 197 Table 1 Instrumental conditions for the determination of aluminium in spices and aromatic herbs by ETA-AASa Step Temperature ᎐ time program for graphite furnace Temperature ŽЊC. Dry Char Atomize Clean 100 150 1700 2500 2800 20 Ramp time Žs. 10 10 10 1 1 1 Hold time Žs. 10 10 20 5 2 2 Gas flow rate Žmlrmin. 300 300 300 0 300 300


a Notes. Wavelength, 309.3 nm; slit width, 0.7 nm; lamp intensity, 25 mA; background correction, deuterium; matrix modifier, 50 ␮ g MgŽNO 3 . 2 ; sample volume, 10 ␮ l; matrix modifier volume, 10 ␮ l; integration time Žpeak area mode., 6 s.

1994. and the use of quartz, polystyrene, polyethylene and Teflon materials is very important ŽPomeranz and Meloan, 1994.. This fact has been carefully considered in this study; blank levels obtained are very low. Disposable polyethylene tips were cleaned as above and then used on the micropipettes. 2.4. Samples A total of 72 samples of 17 different spices and aromatic herbs Ždried or processed. and mixtures, were analysed. Several brand names of each product, representing the most widely accepted and most frequently consumed in Spain, were selected for testing. All samples were commercially available. To optimise the sample amount, preliminary assays were done to ensure that samples were homogeneous and representative of the product investigated ŽPrichard et al., 1996.. The herbs were properly washed in order to remove any additional contamination of Al from the soil. 2.5. Sample mineralization A 0.250-g portion of the homogenized sample was treated with 5 ml of nitric acid and 35 mg of vanadium pentoxide as a catalyst in Pyrex tubes placed in the digestion block and heated at 60ЊC for 30 min and at 120ЊC for 60 min. Solutions were left to cool to room temperature, trans-

ferred to a calibrated flask and diluted to a final volume of 25 ml with bidistilled deionised water. Aluminium was determined in the resulting solution by ETA-AAS. Three portions of each sample were analysed. To validate the proposed method, a microwave digestion bomb was used for acid mineralization of samples, as described Cabrera et al. Ž1994.. Results showed no significant differences between both procedures Ž P - 0.05, Ftest..

2.6. Sample analysis

Aluminium was determined by ETA-AAS in mineralised solutions, under optimised conditions shown in Table 1. The furnace conditions were optimised on the basis of temperature ᎐ time assays for drying, matrix mineralization and atomization of analyte. The use of pyrolyticallycoated graphite tubes with a L’Vov platform, using MgŽNO3 .2 as a matrix modifier gave more reproducible results. Different amounts of MgŽNO3 .2 were tested as the matrix modifier; the best results were obtained with 50 ␮ g for a sample aliquot volume of 10 ␮ l. An ammonium molybdate saturated solution was used to precondition furnace tubes. Optimized assay conditions obviate most matrix interferences and other sources of unspecific absorption. The argon flow was stopped during the atomisation stage in order to increase sensitivity and this did not alter the usable life of


F.F. Lopez ´ et al. r The Science of the Total En¨ ironment 257 (2000) 191 ᎐ 197

Table 2 Statistical analysis of results of Al determination in spices and aromatic herbs by ETA-AAS Sample Cinnamon Paprika
a b

xa 0.031 0.080

b Sn y1

c Sm

x " Sm = t d 0.031 " 1.26 = 10y 3 0.080 " 8.79 = 10y 4

Sm = trx Ž%. 4.07 1.10

1.76 = 10y 3 1.23 = 10y3

5.60 = 10y 4 3.89 = 10y 4

x , mean value of 10 determinations. Sny 1 , standard deviation for 10 replicate determinations. c Sm , standard error of the mean. d The value of Student’s t-test was 2.262 for both samples.

tubes. The same procedure was used for blanks. All of the samples were analysed in triplicate. 2.7. Calibration and analytical characteristics The equation for the calibration graph was ᎏ Absorbance s 4.50 = 10y 3 q 2.98 = 10y 3 ŽAl, ngrml. with r s 0.9997 and ␣ s 1% Ž ␣ , level of significance.. The analytical detection limit we calculated, defined as the concentration of Al corresponding to three times the standard deviation from the blank ŽLong and Winefordner, 1983. was 4.0 pg. The characteristic mass, estimated as the Al amount that produced 0.0044 absorbance units was 10.0 pg. To detect possible interferences and matrix effects, the standard addition method was used with five arbitrarily chosen samples; slopes of the addition calibration graphs for spiked samples were similar to the slope of the calibration graph for the standard solutions. To check the similarity of slopes, the Student’s test was applied ŽCuadros et al., 1995.. The precision of aluminium measurements in the present study was determined by using the procedure described above to measure the Al

content of each of a series of random samples 10 times ŽStiel, 1982.. Results are summarized in Table 2. The accuracy of the method was checked by the recovery percentage for each of series of randomized samples; the Al concentration was determined both without the addition of Al and after various quantities of Al standard were added. The mean recoveries obtained from spiked samples was of 97.90 " 1.20%. Accuracy and precision were also checked in 10 determinations in NBS standard reference material ŽTable 3..

3. Results and discussion The advantages of the analytical method we proposed make it useful for routine analysis, not only for quality control of these products, but also as a measure of environmental contamination and as a tool in assessing medium- and long-term risks to humans. The developed digestion procedure provides rapid, complete mineralization of the sample with minimal losses and external contamination. More than one element at a time can

Table 3 Accuracy and precision of the method against an NBS-certified reference material Reference material Al content a Ž␮ grg. Measured b Certified b Ž%. Accuracy RSD Ž%. Precision

Citrus lea¨ es SRM 1572
a b

91.75 " 5.0

92.0 " 15.0



Dry weight. Mean " S.D. at 95% C.I. interval about the mean Ž n s 10..

F.F. Lopez ´ et al. r The Science of the Total En¨ ironment 257 (2000) 191 ᎐ 197 Table 4 Levels of aluminium Ždry wt.. in spices and aromatic herbs Sample Scientific name n Aluminium Ž␮ grg. Mean Anissed Basil Cinnamon Garlic Bay leaf Mint Mustard Nutmeg Onion Oregano Parsley Paprika Pepper Žblack. Pepper Žwhite. Saffron Thyme Vanilla Mixture Žfor salad. Mixture Žfor meat. Pimpinella anisum Calamintha acinos Cinnamomum acasia Allium sati¨ um Laurus nobilis Menta piperita Brassica juncea Myristica fragans Allium cepa Origanum ¨ ulgare Petroselinum sati¨ um Capsicum annum Piper nigrum Piper nigrum Crocus sati¨ us Thymus ¨ ulgaris Vanilla planifolia ᎐ ᎐ 3 3 6 4 5 4 4 4 3 3 3 6 5 3 4 3 3 3 3 13.35 25.75 28.07 14.89 25.80 22.85 35.30 23.81 14.89 18.36 5.20 24.70 13.34 18.24 18.10 6.50 25.30 18.62 20.80 Range 12.50 ᎐ 14.75 24.80 ᎐ 27.30 18.54 ᎐ 56.50 13.60 ᎐ 15.25 24.70 ᎐ 26.37 22.50 ᎐ 23.00 30.40 ᎐ 38.56 22.81 ᎐ 24.80 14.80 ᎐ 15.20 3.74 ᎐ 40.41 5.10 ᎐ 7.30 11.97 ᎐ 43.61 5.79 ᎐ 24.41 10.49 ᎐ 26.00 14.00 ᎐ 22.24 6.35 ᎐ 7.90 20.80 ᎐ 26.30 14.50 ᎐ 23.59 19.75 ᎐ 22.60


be analysed in a single analytical solution, and the method is suitable for a wide variety of samples. The accuracy and precision of the analytical technique is appropriate for the range of Al found in the experimental samples discussed here. The levels of Al in 72 samples of 17 different spices and aromatic herbs and mixtures used to season salad and meats; the data are given in Table 4. A wide variability of results was observed not only between the different types of spices and aromatic herbs, but also for the same type of sample. The aluminium levels ranged from 3.74 to 56.50 ␮ grg Žexpressed in dry wt..; the presence of Al has been detected in all samples we analysed. The most elevated Al concentrations were found in some samples of cinnamon, mustard, oregano and paprika. Reilly Ž1980. reported data on the Al content in spices with a mean value of 150 mgrkg, and Greger Ž1985. indicated values between 82 and 750 ␮ grg. Wang et al. Ž1991. encountered Al values in dried garlic samples of 58.3 ␮ grg. Muller et al. Ž1998. indicated that most investigated foodstuffs Žvegetables, meat, dairy products. contained - 5 ␮ grg of aluminium Žfresh wt..;

highest concentrations were determined in cocoar cocoa products Ž33 ␮ grg., spices Ž145 ␮ grg., and black tea leaves Ž899 ␮ grg.. These authors indicated that in general, the Al content of frequently consumed food increased in the following order: beverages ) food of animal origin ) food of plant origin. However, several authors reported that the soil Al content is an important factor contributing to total Al concentration in plant tissues; samples taken from plants growing on highAl soils contained higher Al concentrations than similar plants growing on low-Al soils ŽGerhardsson et al., 1994; Eschnauer and Scollary, 1995.. In addition, it is well known that the problem of Al accumulation in plants when these grow in acid soils; the plant growth may be inhibited by Al concentrations of 10 ᎐ 20 ␮ M. Likewise, it was shown that plants absorb this element efficiently from grounds which have alkaline characteristics ŽTyler, 1994.. Abouarab et al. Ž1999. reported the presence of heavy metal in medicinal plants frequently used by both infants and adults Žpeppermint, chamomile, anise, caraway and tilio.; the samples were collected from different sources in the


F.F. Lopez ´ et al. r The Science of the Total En¨ ironment 257 (2000) 191 ᎐ 197 Arruda M, Gallego M, Valcarcel M. Automatic preparation of ´ milk dessert slurries for the determination of trace amounts of aluminium by electrothermal absorption spectrometry. J Anal At Spectrom 1995a;10:55 ᎐ 59. Arruda M, Gallego M, Valcarcel M. Flow-through microwave ´ digestion system for the determination of aluminium in shellfish by electrothermal atomic absorption spectrometry. J Anal At Spectrom 1995b;10:501 ᎐ 504. Berden ´ M, Clarke N, Danielsson LG, Sparen A. Aluminium speciation: variations caused by the choice of analytical method and by sample storage. Water Air Soil Pollut 1994; 72:213 ᎐ 233. Bertholf RL, Wills MR, Savory J. In: Seiler HG, Sigel H, editors. Handbook on toxicity of inorganic compounds. New York: Marcel Dekker, 1988. Cabrera C, Gallego C, Lopez MC, Lorenzo ML, Lillo E. ´ Determination of levels of lead contamination in food and feed crops. J AOAC Int 1994;77:1249 ᎐ 1252. Castro VR. Chromium in a series of Portuguese plants used in the herbal treatment of diabetes. Biol Trace Elem Res 1998;62:101 ᎐ 106. Cuadros L, Garcıa ´ AM, Ales ´ F, Jimenez ´ C, Roman M. Validation of an analytical instrumental method by standard addition methodology. J AOAC Int 1995;78:471 ᎐ 476. Eschnauer HR, Scollary GR. Aluminium in wines. Die WeinWissenschaft 1995;1:24 ᎐ 30. Gerhardsson L, Oskarsson A, Skerfving S. Acid precipitation ᎏ effects on trace elements and human health. Sci Total Environ 1994;153:237 ᎐ 245. Greger JL. Aluminium content of the American diet. Food Technol 1985;39:73 ᎐ 74. Jagannatha KS, Valeswara G. Aluminium leaching from utensils ᎏ a kinetic study. Int J Food Sci Nutr 1995;46:31 ᎐ 38. Kanias GD, Souleles C, Loukis A, Philotheoupanou E. Trace elements and essential oil composition in chemotypes of the aromatic plant Origanum ¨ ulgare. J Radional Nucl Chem 1998;227:23 ᎐ 29. Liu JX, Nordberg GF. Nephrotoxicities of aluminium andror cadmium ᎐ metallothionein in rats, creatinine excretion and metabolism of selected essential metals. Pharmacol Toxicol 1995;77:155 ᎐ 160. Long GL, Winefordner JD. Limit of detection: a closer look at the IUPAC definition. Anal Chem 1983;55:713. Lopez FF, Cabrera C, Lorenzo ML, Lopez MC. Aluminium ´ ´ levels in wine, beer and other alcoholic beverages consumed in Spain. Sci Total Environ 1998;220:1 ᎐ 9. Lopez FF, Cabrera C, Lorenzo ML, Lopez MC. Aluminium ´ ´ content in foods and beverages consumed in the Spanish diet. J Food Sci 2000;65:2. Manning DC, Slavin W, Carnrick GR. Investigation of Al interferences using the stabilized temperature platform furnace. Spectrochim Acta 1982;37B:331 ᎐ 341. Mino Y, Yamada K, Takeda T, Nagasawa O. Metal-containing components in medicinal plants. Chem Pharm Bull 1996;44:2305 ᎐ 2308.

Table 5 Aluminium levels in food of vegetable origin, widely consumed in Spain ŽLopez et al., 1998, 2000. ´ Sample Aluminium content Mean Teaa Ž␮ grg. Camomile teaa Ž␮ grg. Pennyroyal a Ž␮ grg. Coffee a Ž␮ grg. Garlic b Ž␮ grg. Tomatob Ž␮ grg. Carrot b Ž␮ grg. Red wine Žmgrl. Cider Žmgrl.
a b

range 43.42 ᎐ 58.04 55.11 ᎐ 55.40 55.06 ᎐ 58.06 25.60 ᎐ 29.08 5.58 ᎐ 7.50 0.45 ᎐ 0.53 1.68 ᎐ 1.82 0.072 ᎐ 1.25 0.047 ᎐ 0.265

55.46 55.25 56.56 26.98 6.97 0.49 1.75 0.65 0.18

Referred to dry weight. Referred to fresh weight.

Egyptian market. Sathiyamoorthy et al. Ž1997. measured the concentrations of heavy metals ŽZn, Cu, Fe, Mn, Cd, Pb and Ni. in 51 desert plants, of which 42 are considered medicinal, three fodder and six common plant spices. Accumulation of certain heavy metals in some plant species was 2 ᎐ 260-fold higher than the available heavy metal content of the soil of the Negev desert. The interest in metal-containing components of medicinal plants has increased in recent years ŽMino et al., 1996., but studies including aluminium are scarce. We included in Table 5 data on the presence of Al in certain vegetable products, according to other studies we carried out in samples from Spain. Although the values of Al found in the samples of spices and aromatic herbs we analysed do not represent an imminent toxicological risk, periodic determinations are advisable because of the large amounts of these products that are consumed as part of the Mediterranean diet. References
Abouarab AA, Kawther MS, Eltautawy ME, Badeaa RI, Khayria N. Quantity estimation of some contaminants in commonly used medicinal-plants in the Egyptian market. Food Chem 1999;67:357 ᎐ 363. Arruda M, Gallego M, Valcarcel M. Determination of ´ aluminium in slurry and liquid phase of juices by flow injection analysis graphite furnace atomic absorption spectrometry. Anal Chem 1993;15:3331 ᎐ 3335.

F.F. Lopez ´ et al. r The Science of the Total En¨ ironment 257 (2000) 191 ᎐ 197 Minoia C, Sabbioni E, Roncia A, Gatti A. Trace element reference values in tissues from inhabitants of the European community. IV. Influence of dietary factors. Sci Total Environ 1994;141:181 ᎐ 195. Muller M, Anke M, Illiggunther H. Aluminium in foodstuffs. Food Chem 1998;61:419 ᎐ 428. Pomeranz Y, Meloan CE. Food analysis: theory and practice. 3rd New York: Chapman and Hall, 1994. Prichard E, Mackay GM, Points J. Trace analysis: a structured approach to obtaining reliable results. Cambridge: Royal Society of Chemistry, 1996. Reilly C. Metal contamination of food. Londres: Applied Science Publishers, 1980. Sathiyamoorthy P, Vandamme P, Oven M, Gologoldhirsh A. Heavy metals in medicinal and fodder plants of the Negev desert. J Environ Sci Health 1997;32:2111 ᎐ 2123. Slanina P, Frech W, Ekstrom LG, Loof F, Sorach S, Cedergren A. Dietary citric acid enhances absorption of aluminium in antiacids. Clin Chem 1986;32:539 ᎐ 541. Stiel RG. Principles and procedures of statistics. London: McGraw-Hill, 1982.


Storey E, Masters CL. Amyloid, aluminium and the etiology of Alzheimer’s disease. Med J Aust 1995;163:256 ᎐ 259. Taylor A, Branch S, Crews HM, Halls DJ, Owen LM, White M. Atomic spectrometry update ᎏ clinical and biological materials, food and beverages. J Anal At Spectrom 1997; 12:119 ᎐ 221. Tyler G. Plant uptake of aluminum from calcareus soils. Experientia 1994;50:701 ᎐ 703. Vinas I, Hernandez M. Slurry ᎏ ˜ P, Campillo N, Lopez ´ ´ electrothermal atomic absorption spectrometric determination of aluminium and chromium in vegetables using hydrogen peroxide as a matrix modifier. Talanta 1995; 42:527 ᎐ 533. Wang Y, Changhao L, Zhifang X, Wang G, Kuan SS, Rigsby EJ. Determination of aluminium in foods by stabilized temperature platform graphite furnace atomic absorption spectrometry. J Agric Food Chem 1991;39:724 ᎐ 726. Wilplinger M, Schonsleben I, Pfannhauser W. Chromium contents in Austrian foods. Z Lebens Unters Forsch 1995;201:521 ᎐ 523.