WO2001040795A2 - Method for determining that a product has been organically produced - Google Patents

Abstract

A method for determining if an agricultural or horticultural product has been organically produced, comprising the use of elemental analysis of the product by e.g. mass spectrometry such as inductively coupled plasma mass spectrometry (ICPMS), and showing that the elemental content of the assumed organically produced product differs from elemental content data obtained from known conventionally produced products, or similar to elemental content data obtained from known organically produced products.

Description

METHOD FOR DETERMINING THAT A PRODUCT HAS BEEN ORGANICALLY PRODUCED

FIELD OF THE INVENTION

The present invention relates to the field of analysing agricultural and horticultural products Specifically, there is provided a method for determining if such a product has been organically produced The method comprises the use of elemental analysis of the as- sumed organically produced product, e g by mass spectrometry, and the showing that the elemental content of the assumed organically produced product differs from elemental content data obtained from known conventionally produced products, and/or the showing that the elemental content data of the assumed organically produced product are similar to elemental content data obtained from known organically produced products

TECHNICAL BACKGROUND AND PRIOR ART

Organic production is an increasingly important part of the agricultural and horticultural sector Its environmental and economic benefits have captured the attention of many countries, and the consumer demand for e g organically produced food and fibres products, and the demand for more sustainable development, is increasing This demand gives rise to a higher market price for such products than for conventionally produced products In response to the rising demand, such food and fibres products are being placed on the market with indications, e g by labelling, stating or implying to purchasers that they have been produced organically or without the use of artificial fertilisers, pesticides or synthetic chemicals Such indications are applied in order to ensure conditions of fair competition between the producers of products bearing such indications and give the market for organic products a more distinctive profile by ensuring transparency at all stages of production and processing, thereby improving the credibility of such products in the eyes of consumers

However, it is not presently possible to determine if a product has been produced by a conventional production method or has been organically produced Thus, two products identical in all aspect except from the production method, are not easily distinguished, giving rise to possibilities of fraud For this reason, it is not difficult for the unscrupulous to market a conventionally produced product as a genuine organically produced product and thereby obtain an illicit profit when putting this counterfeit product on the market From the consumer's point of view, there can thus be serious doubts with respect to the credibility of products being marketed as organically produced products, which in turn can pose an unfair competition on e g the agricultural sector producing genuine organically produced products

At present, the only way to ensure that products claimed to be organically produced in fact are manufactured in the alleged manner, is to apply and rely on different accreditation and certification systems in order to validate that an "organic" product meets certain national and international standards Such standards are e g disclosed in "Guidelines for the production, processing, labelling and marketing of organically produced foods CAC/GL 32- 1999" (Joint FAO/WHO Food Standards Programme FAO/WHO, Rome, 1999 ), "Council Regulation (EEC) No 2092/91 of 24 June 1991 on organic production of agricultural products and indications referring thereto on agricultural products and foodstuffs", Danish law "Lov om økologisk jordbrugsproduktion nr 363 af 10 Juni 1987', Danish government regulations "nr 697 af 16 juli 2000", "nr 821 af 31 august 2000" and "nr 698 af 16 juli 2000", "Council Regulation (EEC) No 2092/91 of 24 June 1991 on organic production of agricultural products and indications referring thereto on agricultural products and foodstuffs", and in "Council Regulation (EEC) No 1804/99 of 19 July 1999" However, it is primarily up to each individual producer of organic products that these standards and regulations are complied with because the product itself is not controlled

There is thus a need to solve the above problem related to the difficulties in objectively determining that a product has been organically produced

In the prior art several attempts have been made in order to determine the origin and authenticity of different agricultural products

US 5,252,490 discloses a method for identification of the country of origin of cannabis by applying gas chromatographic and mass spectrometπc analysis of marijuana plant material from a country or geographical location, preparing a chemical location profile and comparing a sample of unknown origin with the chemical location profile Also the use of elemental analysis, such as trace element screening, for determining authenticity of geographical origin of e.g. foods and beverages has also been applied in several studies. Thus, Haswell and Walmsley 1998 (Journal of Analytical Atomic Spectrometry, 13:131-134) describe the use of multi-elemental analysis for classification of wines and coffees by applying total reflection X-ray fluorescence analysis, analysing the samples for 11 main elements. By this method it was possible to distinguish each wine type by its country of origin and each coffee type could be grouped by manufacturer.

The elemental uptake in potato tubers which had been conventionally cultivated and fer- tilised with pig slurry or calcium ammonium nitrate was investigated by Bibak et al. (1999). Three different nitrogen levels were included in the study. A discriminant partial least squares regression (DPLS) performed on the calcium ammonium nitrate data and another one performed on the pig slurry data showed a split into three groups represented by the three nitrogen levels in both cases. In this way it was shown that the elemental uptake in potato tubers depends on the nitrogen level, and it was concluded in the study that the pattern of uptakes is dependent on the fertilisation practise.

In the light of the strong influence of nitrogen application on the elemental uptake, the results of Bibak et al. (1999) indicate that it would not be possible to apply multi-elemental analysis as a tool for determining that a product has been organically produced, as it would be impossible to distinguish between organically and conventionally produced products cultivated with pig slurry at the same nitrogen level.

Accordingly, none of the prior art methods known to the present inventors discloses or suggests methods for determining by quantifiable parameters that a product has been organically produced.

According to the present invention, there is now provided a method for determining that a product has been organically produced, and thereby efficiently differentiate conventionally produced products from organically produced products. Thus, there is provided a method for authenticity control of organically produced products. It is an important aspect of the present invention that the method can be used to distinguish between organically produced products and conventionally produced products even when the products have been produced in the same geographical area. SUMMARY OF THE INVENTION

Accordingly, the present invention relates to a method for determining that a product has been organically produced. The method comprises the steps of (a) collecting a sample of the assumed organically produced product, (b) subjecting the sample to an elemental analysis to obtain data for the elemental content, (c) establishing the fact that the product has been organically produced by subjecting said data to a data analysis and showing that the data differs from elemental content data obtained from a conventionally produced product, and/or (d) establishing the fact that the product has been organically produced by subjecting said data to a data analysis and showing that the data are similar to elemental content data obtained from an organically produced product.

DETAILED DISCLOSURE OF THE INVENTION

The primary objective of the present invention is to provide a method for determining that a product has been organically produced, and thus to determine the authenticity and the genuineness the organically produced product.

The term "organically produced" is used herein to designate that a product has been produced by applying an approach to agriculture, horticulture or aquaculture, where the aim is to create integrated, humane, environmentally and economically sustainable production systems wherein reliance on external inputs, whether chemical or organic, is reduced as far as possible.

The key characteristics of agricultural and horticultural products that are "organically produced" include products that are produced by farming methods which comprise (i) ensuring protection of the long term fertility of soils by maintaining organic matter levels, encouraging soil biological activity and careful mechanical intervention; (ii) the provision of crop nutrients indirectly using relatively insoluble nutrient sources which are made available to the plant by the action of soil micro-organisms; (iii) nitrogen self-sufficiency through the use of legumes and biological nitrogen fixation, as well as effective recycling of organic materials including crop residues and livestock manures; (iv) weed, disease and pest control relying primarily on crop rotations, natural predators, diversity, organic manuring, resistant varieties and limited (preferably minimal) thermal, biological and chemical intervention, (v) extensive management of livestock by paying full regard to their evolutionary adaptations, behavioural needs and animal welfare issues with respect to nutrition, housing, health, breeding and rearing, and (vi) careful attention to the impact of the farming system on the wider environment and the conservation of wildlife and natural habitats

More specifically, the term "organically produced" is intended to mean products which are cultivated and/or produced in accordance with the internationally and nationally recognised standards for organically produced products, including the standards set forth in "Guidelines for the production, processing, labelling and marketing of organically produced foods CAC/GL 32-1999" (Joint FAO/WHO Food Standards Programme FAO/WHO, Rome, 1999 ), "Council Regulation (EEC) No 2092/91 of 24 June 1991 on organic production of agricultural products and indications referring thereto on agricultural products and foodstuffs" and as disclosed in Danish law "Lov om økologisk jordbrugsproduktion nr 363 af 10 Juni 1987', Danish government regulations "nr 697 af 16 juli 2000", "nr 821 af 31 august 2000" and "nr 698 af 16 juli 2000", "Council Regulation (EEC) No 2092/91 of 24 June 1991 on organic production of agricultural products and indications referring thereto on agricultural products and foodstuffs", and in "Council Regulation (EEC) No 1804/99 of 19 July 1999"

In contrast hereto, a conventionally produced product is a product which is farmed or produced without the limitations given for an organically produced product Thus, such products are generally produced by e g the reliance on external inputs of both chemical or synthetic organic compounds, such as artificial fertilisers and pesticides

In the present context a "conventionally produced product", which is included in the method according to the invention for the purpose of reference, is a product which is essentially identical to the assumed organically produced product in all aspects except from the production method Thus, it should be understood that the "conventionally " produced reference product and the assumed organically produced product is preferably derived from the same species

Any horticultural, agricultural or aquacultural product may be analysed in accordance with the invention in order to determine if the product has been organically produced Such horticultural products includes fruit and flowers, and vegetables such as onion, pea, cab- bage, sprout, potato, carrot, and salad It is also contemplated that agricultural products of both animal and vegetable origin can be analysed according to the invention Such agricultural products of animal origin includes egg, milk, milk products, meat, meat products, whole animals and parts of animals Agricultural products of vegetable origin includes barely, wheat, rye and oats Analysis of aquacultural products such as fish, shellfish and crustaceans by the method according to the invention is also contemplated and within the scope of the invention

Additionally, it is contemplated that other agricultural products such as fibres products can be efficiently analysed in accordance with the invention Such fibres products includes plant fibres such as cotton, wood fibres, hemp, jute, kapok and sisal, and fibres of animal origin such as wool, silk, cashmere, mohair and alpaca fibres

As mentioned above, the method according to the invention for determining that a product has been organically produced comprises the collection of a sample of the assumed organically produced product Such sample collection can be performed by any available means, e g by manually harvesting the products as described in the below examples

The sample of the assumed organically produced product is subsequently subjected to an elemental analysis to obtain data for the elemental content of the sample In the present context "elemental analysis" is to be understood as an analysis, where the content of at least one or several (including all) elements is determined by chemical or physical methods with or without instruments An element is to be understood as an element of the Periodical Table including different isotopes Typically, the elemental content of at least 10 elements is analysed, such as at least 20, including at least 30, such as at least 40, e g at least 50, including at least 55, such as at least 60 However, it is also contemplated that in certain embodiments of the invention, it might be sufficient to analyse the elemental content of a relatively low number of elements, e g within the range of 1-9 elements In a presently preferred embodiment of the invention the elemental content of at least 63 ele- ments is analysed

It will be appreciated, that it is also within the scope of the invention that further analytical data other than elemental content data for the assumed organically produced product can be applied in accordance with the invention Accordingly, it is contemplated that analytical data for chemical compounds, such as proteins, carbohydrates, lipids, aromatic com- pounds, vitamins, enzymes, secondary metabolites, i.e. alkaloids, can advantageously be incorporated in the method according to the invention. Such analytical data includes e.g. concentration data and activity data. It is also contemplated that further analytical data such as measurements of physical data of the sample of the assumed organically pro- duced product or characteristic parts of the sample, may be applied in accordance with the invention. Such analytical data may be derived from measurements of e.g. dry mass, humidity/water content, electrical conductivity, pH, mass and volume of the whole sample or characteristic parts of the sample like yolk and white of egg and the core in carrot.

In accordance with the invention, the elemental analysis can be performed by any suitable means including analytical instruments for determining elemental concentrations such as a mass spectrometer. By applying mass spectrometry (MS) it is possible to use the difference in mass-to-charge ratio (m/z) of ionised atoms or molecules to separate them from each other according to their masses. Presently available mass spectrometers includes high resolution inductively coupled plasma mass spectrometers (HR-ICPMS), quadrupole inductively coupled plasma mass spectrometers (Q-ICPMS), ion cyclotron resonance mass spectrometers, ion trap mass spectrometer, time-of-flight mass spectrometers, direct reader mass spectrometers, multiple-collector magnetic mass analysers, twin- quadrupole instruments, and collision cell interfaces.

In a presently preferred embodiment a high resolution inductively coupled plasma mass spectrometer (HR-ICPMS) is applied. An inductively coupled plasma (ICP) is a very high temperature (7000-8000K) excitation source that efficiently desolvates, vaporises, atomises, and ionises atoms.

It is also within the scope of the invention that the elemental analysis can be performed by use of optical spectroscopic analysis. Optical spectroscopy is the use of the absorption, emission, or scattering of electromagnetic radiation by atoms or molecules (or atomic or molecular ions) to qualitatively or quantitatively study the atoms or molecules, or to study physical processes. In one embodiment of the invention an optical spectrometer is applied for the elemental analysis. Presently available optical spectrometers includes: inductively coupled plasma optical emission spectrometers (ICP-OES, ICP-AES), inductively coupled plasma atomic fluorescence spectrometers (ICP-AFS), flame atomic absorption spectrometers (AAS), flame atomic emission spectrometers (AES), electrothermal atomisation atomic absorption spectrometers (GFA), X-ray fluorescence spectrometer (XRF), and total reflection X-ray fluorescence spectrometers (TXRF)

A detailed description of inductively coupled plasma mass spectrometers (ICPMS), HR- ICPMS, Q-ICPMS, ion cyclotron resonance mass spectrometers, an ion trap mass spectrometers, time-of-fhght mass spectrometers, direct reader mass spectrometers, multiple- collector magnetic mass analysers, twin-quadrupole instruments, inductively coupled plasma optical emission spectrometer (ICP-OES, ICP-AES) and inductively coupled plasma atomic fluorescence spectrometers (ICP-AFS) can be found in Montaser (1998), and a description of flame atomic absorption spectrometers (AAS), flame atomic emission spectrometers (AES), and electrothermal atomisation atomic absorption spectrometers (GFA) are described in Skoog et al (1992) Total-reflection x-ray fluorescence spectrometry (TRXF) is described in Klockenkaemper (2000)

The choice of technique will often depend upon the levels of the elements of interest and the nature of the sample, and if appropriate a preconcentration of the sample can be applied in order to achieve the desired limit of detection

In the present context, the elemental content of a sample is the quantitative content of an element given as the concentration expressed as the mass of the element in one mass unit of sample material, e g in μg/kg sample material The concentration may be expressed in mass derived dimensions of the element in one unit of mass or volume derived dimension of the sample material

It is, as mentioned above, an important aspect of the present invention that the method can be used to discriminate between organically produced products and conventionally produced products even when the products have been produced in the same geographical area Thus, it should be understood that e g vegetables cultivated in a field separated into two areas, wherein the vegetables in one of the areas are cultivated by applying or- ganic cultivation principles and the vegetables in the other area are cultivated by the use of conventional cultivation principles, can be distinguished by the method according to the invention

Likewise, it is also an important aspect that the method of the invention can e g be used to distinguish a conventionally produced agricultural product of animal origin from a or- ganically produced product derived from essentially the same species It is also contemplated that it is possible to make this distinction even if the animals from which the conventionally and the organically produced products are derived, have been feed with conventionally and organically produced fodder, respectively, containing essentially the same amounts and proportions of energy-giving nutrients such as lipids, carbohydrates and proteins

As is mentioned above, an important step of the method according to the invention is to establish the fact that the product has been organically produced by subjecting the ele- mental content data of the sample of the assumed organically produced product obtained from the elemental analysis to a data analysis, and showing that these data differs from elemental content data obtained from known conventionally produced reference products

Optionally, the fact that the product has been organically produced can be established by applying a known organically produced product as reference, by subjecting the elemental content data of the assumed organically produced product obtained from the elemental analysis to a data analysis, and showing that these data are similar to elemental content data obtained from known organically produced reference products

It should also be understood that the fact that the product has been organically produced can be established by applying both known organically produced products and a known conventionally produced products as reference products

In the present context, the elemental content data obtained from known conventionally or organically produced products is to be understood as the elemental content of conventionally or organically produced products obtained essentially by the same method as for the assumed organically produced product Thus, it should be understood that these elemental content data obtained from known conventionally and organically produced products are used as reference data

In one embodiment of the invention, these reference data can be obtained at the same time as the data for the assumed organically produced product, which is also described in the below example However, it is also contemplated that it would be advantageous to store the elemental content data for the known conventionally or organically produced products in a database, and to use these data as references in the method according to the invention

In one useful embodiment, the above database could e g contain elemental content data for different agricultural products of vegetable origin categorised according to their geographical origin In another useful embodiment, the database could e g include elemental content data for different agricultural products of animal origin categorised according to the animal species from which they are derived

The data analysis of the elemental content of the assumed organically produced product and the showing that these data differs from or are similar to elemental content data obtained from a known conventionally or organically produced product, can be performed by manual selection such as by the judgement of a skilled person with the aim to find obvious differences between the elemental content data

However, in a more preferred embodiment of the invention, the data analysis and the showing that the data differs from data obtained from known conventionally produced products, or are similar to data obtained from known organically produced products, is performed by statistical analysis Such statistical analysis includes multivariate analysis including principal component analysis and partial least square regression and methods derived from these methods

Multivariate analysis, which is well known in the art, is a statistical method of using many variables to forecast, predict, or understand a situation Multivariate analysis gathers and puts together all possible information on numerous variables to make predictions and answer questions An example of multivariate analysis is principal component analysis (PCA), which is a projection method helping to visualise the information in a data table (X) The purpose of PCA is to express the main information in the X-vaπables by a lower number of latent variables (referred to as principal components)

Another example of multivariate analysis which is applicable for the purpose of the present invention, is discriminant partial least squares regression (DPLS) This analysis method is used to find latent variables in the X space which have a maximum covaπance with the Y-vanables (the dependent variables) Thus, linear combinations of the X- variables (the independent variables) are found that are tilted to have maximum prediction ability for the Y-variable(s).

Other presently available multivariate analysis methods includes: Analysis of variance (ANOVA), Linear discriminant analysis (LDA), Soft independent modelling of class analogy (SIMCA), K-nearest neighbours (KNN), Cluster analysis (CA), Canonical Variates Analysis (CVA) (also referred to as Discriminant Function Analysis (DFA)), Artificial neural networks (ANNs), Factor analysis, Multidimensional scaling (MDS), and Correspondence factor analysis (CFA). The above methods are all described in the literature, e.g. Brown et al. (1992), Massart et al. (1988), Sharaf et al. (1986), Box et al. (1978), Coomans et al. (1979), Lachenbruch (1975), Wold et al. (1987), Massart et al. (1988), Martens et al. (1989), Sjόstrόm et al. (1986), Coomans et al. (1982), Manly (1994), Massart et al. (1983), Manly (1994), Haykin (1994), Malinowski (1991), Piggott (1986), and Mellinger (1987).

It is also contemplated that other suitable multivariate analysis methods for the present invention can be developed or derived from the above mentioned multivariate analysis methods.

The invention will now be explained in further details in the following, non-limiting exam- pies and the drawings wherein:

Figure 1 shows the score plot of discriminant partial least square regression (DPLS) of concentrations of 24 elements in onions. The concentrations are mean values of ten samples from each of 11 conventionally and 10 organically cultivated sites. Organically pro- duced onions are marked with second letter "E" and conventionally produced onions are marked with second letter "C".

Figure 2 shows a loading plot of discriminant partial least square regression (DPLS) of organically and conventionally produced onions corresponding to Figure 1.

Figure 3 shows the score plot of principal component analysis (PCA) of concentrations of 63 elements in 210 individual samples of onions from 11 conventionally and 10 organically cultivated sites. Organically produced onions are marked with second letter "E" and conventionally produced onions are marked with second letter "C". Figure 4 shows the scatter plot of loadings for the first and third principal component in the PCA model for onions corresponding to Figure 3

Figure 5 shows the score plot of discriminant partial least square regression (DPLS) of concentrations of 14 elements in potato tubers The concentrations are mean values of twelve samples from each of 12 conventionally and 12 organically cultivated sites Organically produced potatoes are marked with second letter "E" and conventionally produced potatoes are marked with second letter "C"

Figure 6 shows a loading plot of discriminant partial least square regression (DPLS) of organically and conventionally produced potatoes corresponding to Figure 5

EXAMPLE 1

ANALYSIS OF ORGANICALLY AND CONVENTIONALLY PRODUCED ONIONS

Sampling and sample preparation

Onions were collected from eleven conventionally cultivated farms and ten organic cultivated farms from Funen and mid-Jutland in Denmark, and thus essentially from the same geographical area The sites were selected to avoid contamination from human activities not included in the agricultural practice

To avoid contamination of samples with the elements, laboratory modification, sampling device, special equipment for sample preparation, laboratory ware and cleaning procedure were used in accordance with procedures outlined below

The onions were harvested manually and allowed to weather for one week on soil sur- face Ten undamaged, healthy, average sized and normal shaped onions (Allium cepa Hysam) were sampled evenly across each site All onion samples were collected using Nitπlite gloves (Nitπlite, powder free, Ansell Edmont) and loose soil and stalk were shaken off, thereafter placed in a polyethylene tissue bag with closing tape The collected onions were stored in a drying room with shelter from the rain but open from the wind to maintain natural drying conditions, where they were dried for a period of more than two weeks This drying procedure is similar to normal drying procedure for basket onion in Denmark In order to eliminate or at least minimise the risk of contamination, all sample preparations were performed under controlled condition in three rooms with lock-gate connection The rooms are classified R1 (ordinary condition), R2 (fairly clean) and R3 (clean, class 1000 room)

All sample preparations subsequent to the cleaning procedures were carried out in class 1000 environment (R3) Disposable surgical latex gloves (Gammex, sterile and powder free, Ansell Edmont) and full laboratory dress (TyvekTM) were worn throughout the procedure Laboratory wares were stored in a clean air environment (R3)

In R1 the onion sample (1 onion = 1 sample) was cut from both ends (approx 1 mm) with a nitride hardened titanium knife on polycarbonate carving board, then the outer layer was peeled off and all 10 samples from one site were placed in a polyethylene terephthalate (PET) bag The sample was then passed through a lock into R2 The above described R1 procedure was repeated in R2 and R3 In R3, the onion sample was quartered and homogenised in a blender (EVA, type 267732, DK) modified with a nitride hardened titanium cutter

Multi-element determination

The onion samples were digested with redistilled nitric acid (Merck p a subboiled in R3) in a microwave oven (MDS 2000, CEM Co , Matthews, NC, USA) equipped with 12 closed Teflon PFA (per fluoro alkoxy) digestion vessels (CEM Co , Matthews, NC, USA) From the homogenised sample of one onion a mass of approximately 2 g was accurately weighed (to the nearest 0 0001 g) into each vessel and 10 0 ml of redistilled nitric acid (Merck p a subboiled in R3) were added The microwave oven (power level 504 W) was programmed to run at increasing pressure of 40, 85, and 175 psi in 3 steps The pressure was held constant for 3, 3 and 5 minutes during the 3 steps, respectively

The clear light yellow digest without any residue was then cooled to room temperature and transferred quantitatively to a 50-ml polyethylene flask, and double deionised water (see description below) was added to a mass of approximately 30 g (weighed to the near- est 0 0001 g) These sample solutions were stored at 5°C until analysis 2 g (weighed to the nearest 0.0001 g) of the sample solution was diluted with double deionised water to 10 g (weighed to the nearest 0.0001 g) for the HR-ICPMS measurement. Each batch consisted of 10 samples, a reagent blank and a reference material (Leek In-house material from the National Food Agency of Denmark). The reagent blank was used to evaluate contamination from digestion vessels and reagents, whereas the reference material was used to check the efficiency of the digestion to secure uniformity of the sample solution matrices from batch to batch.

The precision and accuracy of the analytical method were determined by 10 subsamples from one onion and 10 samples of reference material, respectively. Element concentrations for some of the samples were estimated in analytical combination with different spiked samples to verify that the sample solution matrix is similar for different onions, when the digestion condition is the same.

HR-ICPMS (PlasmaTrace2, Micromass, UK) was used to determine 63 elements (Ag, Al, Au, B, Ba, Be, Bi, Ca, Cd, Ce, Co, Cr, Cs, Cu, Dy, Er, Eu, Fe, Ga, Gd, Ge, Hf, Hg, Ho, In, Ir, K, La, Li, Lu, Mg, Mn, Mo, Na, Nb, Nd, P, Pb, Pr, Pt, Rb, Re, Ru, S, Sb, Sc, Si, Sm, Sn, Sr, Tb, Te, Th, Ti, Tl, Tm, U, V, W, Y, Yb, Zn, and Zr) in the onion samples. The above elements are the elements that we found possible to measure using a routine HR-ICPMS method. The plasma conditions and acquisition parameters are summarised in Table 1.

The isotopes chosen for analysis are chosen in order to obtain best possible sensitivity (highest natural abundance) and to avoid overlap from polyatomic interference's. Where overlap from polyatomic ions could not be avoided a higher resolution (defined as m/Δm) of either 4000 or 10000 was applied to resolve the analyte peaks from the otherwise interfering polyatomic peaks e.g. ⁵⁶Fe⁺ can be separated from ⁴⁰Ar¹⁶O⁺ (Δm = 0.0049 amu) using a resolution of 4000. When a higher resolution is applied the ion transmission is decreased significantly at resolution 4000 and 10000 the transmission is approximately 15 % and 1 %, respectively.

During the method development also arsenic and selenium was measured, but it was not possible to set-up a reliable method for the determination of these two elements in onions. Even though the arsenic signal (⁷⁵As⁺) can be separated from the interfering argonchlorid signal (⁴⁰Ar³⁵CI⁺) using a resolution of 10000, the content of arsenic in the onions were to low to be measured reliable given the low ion transmission at resolution 10000. For selenium the major isotope (⁸⁰Se⁺) is interfered by the argondimer (⁴⁰Ar²⁺) which can not be resolved using even high resolution, one of the less abundant selenium isotopes has to be chosen for analysis (e.g. ⁸²Se⁺). The ⁸²Se isotope can be resolved from most interference's using a resolution of 4000, but the signal can not be resolved from ⁸²Kr⁺. Kr is an impurity in the argon gas. Given the surprisingly low levels of selenium present in the onions it was not possible to obtain reliable selenium concentrations given the relatively low ion transmission and the interference from Kr.

The quantification was performed using standard addition calibration in order to eliminate interference's from the sample matrix. Standard addition calibration was carried out by the addition of 6 multi-element standard solutions (Perkin-Elmer). Each standard was added at 3 concentration levels to separate samples. The analyses were performed in a class 1000 room (R3). One standard addition calibration curve was obtained for each element for every 20 samples of onions assuming a similar sample solution matrix for all onion samples digested under the same conditions. The efficiency of the digestions was checked for each batch of samples (10 samples of onions from the same site) by the reference material samples in the batches.

Table 1 Instrumental Conditions for the PlasmaTrace 2 (HR-ICPMS)

RF power/w 1350

Gas flow rates/I mm^" Plasma 12 5

Auxiliary 1 5

Nebu ser 0 920 - 0 980 (adjusted daily to maximum signal intensity)

Sample uptake rate/ml mm^" 0 7 Ion sampling depth Adjusted to maximum signal intensity Ion lens settings Adjusted to maximum signal intensity Sampling cone Nickel, 1 0 mm orifice diameter Skimmer cone Nickel, 0 7 mm orifice diameter

Acquisition parameters

Resolution 400 Peak widths 3

Points/peak width 20

Dwell time 10 ms

Scans 2 (⁷Lι, ⁹Be, ¹¹B, ⁷⁹Br, ⁸⁵Rb, ⁸⁸Sr, ⁸⁹Y, ⁹⁰Zr, ⁹³Nb, ⁹⁸Mo, ¹⁰⁷Ag, ¹¹¹Cd, ¹¹⁵ln, ¹¹⁸Sn, ¹²¹Sb, ¹²⁶Te, ¹³³Cs, ¹³⁸Ba, ¹³⁹La, ¹⁴⁰Ce, ¹⁴¹Pr, ¹⁴⁶Nd, ¹⁷⁸Hf, ¹⁸²W, ¹⁹⁵Pt, ¹⁹⁷Au, ²⁰²Hg, ²⁰⁵TI, ²⁰⁸Pb, ²⁰⁹Bι, ²³²Th) (¹⁰¹Ru, ¹⁵⁷Gd, ¹⁵⁹Tb, ¹⁶³Dy, ¹⁶⁵Ho, ¹⁶⁶Er, ¹⁶⁹Tm, ¹⁷²Yb, ¹⁷⁵Lu, ¹⁸⁵Re, ¹⁹³lr, ²³⁸U)

Resolution 4000 Peak widths 3

Points/peak width 20

Scans 2

Dwell time (ms) 10 (²⁴Mg, ⁸Sι, ³¹P, ³⁴S, ⁴⁴Ca, ⁴⁸Tι)

20 (⁴⁵Sc, ⁵¹V, ⁵²Cr, ⁵⁵Mn, ⁵⁶Fe, ⁵⁹Co, ⁶⁰Nι,

⁶³Cu, ⁶⁶Zn, ⁶⁹Ga) 30 (²⁷AI)

Resolution 10,000 Peak width 5

Points/peak width 20

Dwell time 10 ms,

2 scans ²³Na, ³⁹K

Dwell time 40 ms,

1 scan ¹⁵³Eu

Dwell time 50 ms,

2 scans ⁷²Ge Cleaning Procedures

All labware was washed with 20% citric acid, 10% nitric acid, deionised water (DW with measured resistance > 10 MΩ cm, Millipore system) and double deionised water (DDW with resistance of 18 2 MΩ cm, Elgastat, UHQPS) and then air dried in the clean room (R3)

The Teflon vessels were washed with 10% nitric acid, DW and DDW between each digestion All the sample preparation equipment that has been in contact with the samples were washed with DW and DDW and air dried in the clean room (R3)

Data processing and statistical analysis

A computer program was developed in Risoe's Engineering and Computer Department which was capable of handling standard additions with different multi-element standard solution in one procedure The integrated data from the HR-ICPMS instrument, dilution factors and multi-element standard concentrations were loaded into the program and the concentrations of the elements in each sample were calculated

The results were analysed statistically using simple correlation coefficients (r) in the Stat- graphics statistical software package (Statgraphics Pluse, 1995) and multivariate analysis was performed with Unscrambler software package (CAMO ASA, 1996)

Results and discussion

The concentration of 63 elements in onion taken from 21 background areas are shown in Table 2

TABLE 2

Onions (Allium cepa Hysam)

Conventionally produced (μg/kg) Organicεilly produced (μg/kg)

Element p- n Mean Median Sdev Confin Mean Median Sdev Confivalue dence dence (%) interval interval

Na23 5059 98 210000 170000 110000 22000 94 160000 140000 54000 11000

Mg24 537 106 100000 100000 17000 3200 96 110000 110000 18000 3600

Si 28 2969 97 11000 8800 7100 1400 84 9000 8200 3400 720

P31 6596 108 440000 440000 75000 14000 94 420000 410000 80000 16000

S34 8152 108 120000 1300000 480000 90000 95 12000001200000 540000 110000

K39 5841 105 160000 1600000 940000 180000 90 180000013000001200000 240000

Ca44 895 98 200000 180000 84000 17000 91 140000 130000 56000 11000

Li 7 1788 89 064 b>c 15 03 72 b>c b>c 06 014

Be 9 1725 86 0033 0024 0033 00069 78 0062 004 0065 0015

B11 156^* 102 1600 1700 810 160 96 890 910 320 65

AI27 672 98 250 150 240 47 96 130 110 80 16

Sc45 483^* 97 17 15 093 018 92 11 09 07 014

Tι48 047* 99 910 740 560 110 96 390 320 260 51

V51 3793 82 044 043 015 0033 83 053 044 031 0067

Cr52 8898 102 96 83 63 12 90 92 79 43 089

Mn55 8637 102 1600 1500 620 120 95 1500 1500 430 86

Fe56 736 101 2600 2600 800 160 95 3200 3000 750 150

Co 59 945 100 18 15 11 021 93 13 1 094 019

Cu63 1034 102 500 510 140 28 96 620 620 160 31

Zn66 948 101 3400 2900 1400 280 95 3500 3300 1200 230

Ga69 6128 99 022 014 026 0052 74 017 007 025 0056

Ge72 5884 62 11 98 59 15 77 95 98 44 099

Rb85 007* 102 380 330 170 33 95 670 650 210 42

Sr88 09^* 102 910 780 510 99 94 550 460 260 53

Y89 265* 99 022 018 015 0029 95 011 0092 0063 0013

Zr90 5446 82 31 23 25 053 84 34 24 3 064

Nb93 8067 90 14 11 14 028 90 12 092 098 02

Mo 98 3204 78 8 66 8 18 87 21 68 39 83

Ru101 3869 99 016 011 021 0041 95 011 0079 015 003

Ag107 4775 100 065 056 035 0069 92 059 056 029 0059

Cd 111 1522 100 22 18 14 27 96 15 12 10 2

In 115 381 98 042 033 034 0068 92 03 013 049 01 TABLE 2 (continued)

Onions (Allium cepa Hysam)

Conventionally produced (μg/kg) Organically produced (μg/kg)

Element P- n Mean Median Sdev Confin Mean Median Sdev Confivalue dence dence

(%) interval interval

Sn118 8926 98 35 36 17 033 92 36 31 21 043

Sb121 268^* 102 18 17 1 02 95 1 091 062 012

Te126 3654 83 22 23 092 02 86 19 19 11 023

Cs133 1878 81 021 012 034 0074 84 055 045 064 014

Ba138 7639 102 110 79 89 17 96 100 87 64 13

La 139 2372 85 021 012 021 0044 85 013 011 0094 002

Ce140 1187 100 23 18 23 046 93 067 067 17 035

Pr141 693 101 0034 0015 0057 0011 95 0023 0011 0041 00082

Nd 146 137 101 017 01 02 0039 95 0098 0082 0083 0017

Sm147 6223 102 0044 0039 0045 00088 95 0038 0018 0052 001

Eu153 2063 82 032 015 061 013 84 014 b>c 05 011

Gd 157 633 101 005 0036 0054 0011 95 0085 0058 0078 0016

Tb159 112* 102 00069 00049 0011 00021 96 00014 b>c 00082 00016

Dy 163 051* 102 0017 0011 0026 0005 96 0046 0037 0047 00094

Ho 165 9482 102 00025 00013 0005000096 96 00023 b>c 00053 00011

Er166 8037 102 000065 b>c 002 00039 95 b>c b>c 0026 00053

Tm169 33 102 00048 00039 00042000081 96 00041 000063 00057 00011

Yb172 3247 102 00048 000046 0012 00023 94 00022 b>c 0014 00029

Lu 175 882 102 b>c b>c 00061 00012 96 00017 b>c 00086 00017

Hf 178 3136 71 084 031 15 035 82 043 022 047 01

W182 4747 64 17 15 79 19 88 19 19 82 17

Re 185 4064 98 0014 00073 0028 00055 91 00089 00081 0022 00046

Ir 193 1992 93 0034 0018 0061 0012 86 0012 00003 0032 00068

Pt195 5066 88 02 01 029 0061 86 015 01 016 0033

Au 197 696 91 13 082 15 032 93 068 056 037 0075

Hg202 6739 72 12 13 3 068 85 12 12 39 083

TI205 5924 101 07 047 057 011 94 078 069 041 0082

Pb208 8703 99 62 54 36 072 94 6 53 33 067

Bι209 032^* 100 0086 008 0051 001 90 016 015 0086 0018

Th232 3529 98 11 081 087 017 84 13 062 18 038

U238 347^* 90 0047 0031 0047 00098 89 0022 0019 0017 00036 Outliers were omitted from the data material. For some elements it was necessary to remove 1-5 estimations after spiking on high levels due to tailing. Therefore the number of samples (n) in Table 2 are different. The detection limits (μg/kg) are taken as 3 times the standard deviation of 10 replicates of the blank determination. The elemental concentra- tion within the sites is uniform. The mean, median, and 95% confidence interval values in Table 2 are calculated from all samples from the 21 sites.

For most of the elements the mean exceeds the median (Table 2) because a few sites have very high values. Some elements have concentration levels below the mean of 10 consecutive blank determination, these are listed as b > c in Table 2. In these cases the negative values (c - b) are included in the statistical calculations. 11 elements have P- values less than 5% and these elements significantly separate organic produced onions from conventionally produced.

The score plot of a discriminant partial least square regression (DPLS) of normalised mean values of 24 elements is shown in Figure 1. Elements only participating as noise in the regression are excluded.

It is clearly seen from the score plot in Figure 1 , that the onions are separated in two dis- tinct groups: conventional an organically produced onions wherein the group with second letter "E" is the organic cultivated onions and the other group marked with second letter "C" is the conventionally cultivated onions. The 24 elements used in order to separate the onions are shown in the loading plot in Figure 2.

Additionally, principal component analysis (PCA) was applied to the 63 elements measured in the individual onion samples from the 21 farmers to investigate the relevant and interpretable structure in the data. The data set consisted of a table containing the results of the elemental analysis performed on the 204 samples; i.e. 204 objects and 63 variables. The variables were weighted with the inverse of the standard deviation of all ob- jects before the PCA. This was done to compensate for the different scales of the variables. It appears from Figure 3, that the onion samples split up into groups according to the farming method when the scores of the first and third principal components (PCs) are plotted against each other. In Figure 3 the samples named FE, GE, HE, IE, JE, LE, ME, NE, OE, and EE represent organically grown onions and the samples named PC, QC, RC, SC, TC, UC, VC, XC, YC, and ZC represent conventionally grown onions. The 63 elements used in order to separate the onions are shown in the scatter plot of the loadings for PC1 and PC3 in Figure 4 The model is based on individual samples from each farmer and thus includes the variations between the individual samples from each farmer In case of authority control of onions, this would be an advantage because it would be possible to identify even a few conventionally grown onion samples in a batch of organically grown onion samples A PCA model based on mean values for each farmer was not found to improve the separation of the two farming methods Models based on a reduced number of elements did not improve the separation either

EXAMPLE 2

ANALYSIS OF ORGANICALLY AND CONVENTIONALLY PRODUCED POTATOES CULTIVATED ON THE SAME GEOGRAPHICAL LOCATION

Two fields located one kilometre apart at Risø National Laboratory in Roskilde, Denmark were both divided into two plots Potatoes (Solanum tuberosum, Folva) were grown in all plots One plot at each location was cultivated organically and one plot at each location was cultivated conventionally with aid of artificial fertilisers and pesticides Each plot was further divided into 3 subplots fertilised with 0, 60, and 120 kg of N/ha Conventionally cultivated plots were fertilised with calcium ammonium nitrate and organically cultivated plots were fertilised with pig slurry

Sampling and sample preparation

A total of 60 undamaged, healthy, and average-sized potato tubers from 12 potato plants (five tubers per plant) were collected by hand digging in an even spread over each potato subplot All potato samples were collected with Nitnle gloves (Nitnlite, powderfree, Ansell Edmont) Tuber cultivar was noted, and tubers were placed in polyethylene tissue bags with closing tape and transported to the laboratory for analysis

Laboratory modification, special equipment for sample preparation, digestion, laboratory ware cleaning procedures, and deionised water (DW) and double deionised water (DDW) supply are described in Example 1 The potatoes collected from a field were stored 6-7 weeks at 5°C and 99% humidity and then prepared for analysis as described below.

Sample preparation was performed under controlled conditions in three rooms with lock- gate connection. The rooms are classified as: R1 (ordinary condition), R2 (fairly clean), and R3 (clean, class 1000 room). Five tubers from one potato plant (representing one sample) were rinsed in tap water and then scrubbed gently in deionised water using a soft nylon brush to remove adhering soil. The washing procedure in deionised water was performed in R1 and was repeated twice. The tubers were then passed through a lock into R2 and the above described procedure was then repeated with deionised water and twice with double deionised water under fairly clean laboratory conditions. The washed and clean tubers in R2 were passed through a lock into R3. In R3, the 5 tubers were cut from both ends (approx. 10-15 mm) with a nitride hardened titanium knife on a polycarbonate carving board. Then a 20 mm (diameter) longitudinal cylindrical piece was taken from each tuber with a sharp nitride hardened titanium tube which was pressed through the tuber. The pieces (pure pulp) were then homogenised in the blender (EVA, type 267732, DK) modified with a nitride hardened titanium cutter. Disposable latex gloves (Gammex, sterile, powder free, Ansell Edmont) and full laboratory dress (Tyvek™) were worn throughout the procedure.

Multi-element determination

Approximately 1.5 g of the homogenised material from each sample was accurately weighed (to the nearest 0.0001 g) into each digestion vessel and 10 ml of redistilled nitric acid (Merck p. a. subboiled in R3) was added. The samples were digested in a microwave oven (MDS 2000 CEM Co., Matthews, NC, U.S.A.) equipped with 12 closed Teflon PFA (per fluoro alkoxy) digestion vessels (CEM Co., Matthews, NC, U.S.A.). The microwave oven was programmed to run at increasing pressure at 2.8, 5.8 and 12 bar in 3 steps. The pressure was held constant for 3, 3 and 5 minutes during each of the 3 steps, respec- tively. The clear, light yellow, residue-free digest was then cooled to room temperature and transferred quantitatively to a 50-ml polyethylene flask, and double deionised water was added to a final mass of approximately 30 g (weighted to the nearest 0.0001 g). These sample solutions were stored at 5°C until analysis. Two grams (weighed to the nearest 0.0001 g) of the sample solution was diluted with double deionised water to 10 g (weighed to the nearest 0.0001 g) for HR-ICPMS measurement. HR-ICPMS (PlasmaTrace2, Micromass, UK) was used to determine 58 elements (Ag, Al, Au, Ba, Be, Bi, Ca, Cd, Ce, Co, Cr, Cs, Cu, Dy, Er, Eu, Fe, Ga, Gd, Ge, Hf, Ho, In, Ir, La, Li, Lu, Mn, Mo, Nb, Nd, P, Pd, Pr, Pt, Rb, Re, Rh, Ru, S, Sb, Sc, Si, Sm, Sn, Sr, Ta, Tb, Te, Th, Tl, Tm, U, V, Y, Yb, Zn, and Zr) in the potato samples. The above elements are those we found possible to measure using a routine HR-ICPMS method. The plasma conditions and acquisition parameters are summarised in the below Table 3.

Table 3. Instrumental Conditions for the PlasmaTrace 2 (HR-ICPMS)

RF power/w 1350

Gas flow rates/I mm ¹ Plasma 12 5

Auxiliary 1 5

Nebuliser 0 0 9922C0 - 0 980 (adjusted daily to maximum signal intensity)

Sample uptake rate/ml mm^"1 0 7 Ion sampling depth Adjusted to maximum signal intensity Ion lens settings Adjusted to maximum signal intensity Sampling cone Nickel, 1 0 mm orifice diameter Skimmer cone Nickel, 0 7 mm orifice diameter

Acquisition parameters

Resolution 400 Peak widths 3 Points/peak width 20 Dwell time 10 ms Scans 2 (⁸⁵Rb, ⁸⁸Sr, ⁸⁹Y, ⁹³Nb, ⁹⁸Mo, ¹⁰⁷Ag, ¹¹¹Cd, ¹¹⁵ln, ¹¹⁸Sn, ¹²¹Sb, ¹³³Cs, ¹³⁸Ba, ¹³⁹La, ¹⁴¹ Pr, ¹⁴⁶Nd, ¹⁴⁷Sm, ¹⁹⁵Pt, ¹⁹⁷Au, ²⁰⁵TI,

208_pb 209_{B| |} 232_Th)

1 (¹⁰¹Ru, ¹⁵⁷Gd, ¹⁵⁹Tb, ¹⁶³Dy, ¹⁶⁵Ho, ¹⁶⁶Er, ¹⁶⁹Tm, ¹⁷²Yb, ¹⁷⁵Lu, ¹⁸¹Ta, ¹⁸⁵Re, ¹⁹³lr,

238

U)

Resolution 4,000 Peak widths 3

Points/peak width 20

Scans 2

Dwell time (ms) 10 (³¹P, Ca, ⁴⁸Tι) 20 (⁴⁵Sc, ⁵¹V, ⁵²Cr,

⁶⁶Zn, ⁶⁹Ga) 30 (²⁷AI)

Resolution 10,000 Peak width 5

Points/peak width 20

Scans 2

Dwell time (ms) 100 (¹⁰³Rh, ¹⁰⁵Pd) The isotopes chosen for analysis are chosen in order to obtain best possible sensitivity (highest natural abundance) and to avoid overlap from polyatomic interference. Where overlap from polyatomic ions could not be avoided a higher resolution (defined as m/Δm) of either 4000 or 10000 was applied to resolve the analyte peaks from the otherwise in- 5 terfering polyatomic peaks e.g. ⁵⁶Fe⁺ can be separated from ⁴⁰Ar¹⁶O⁺ (Δm = 0.0049 amu) using a resolution of 4000. When a higher resolution is applied the ion transmission is decreased significantly at resolution 4000 and 10000 the transmission is approximately 15% and 1%, respectively. The quantification was performed using standard addition calibration in order to eliminate interference from the sample matrix. Standard addition calibra- 0 tion was carried out by the addition of 6 multi-element standard solutions (Perkin-Elmer). Each standard was added at 3 concentration levels to separate samples. The analyses were performed in a class 1000 room (R3). One standard addition calibration curve was obtained for each element for every 24 samples of potatoes assuming a similar sample solution matrix for all potato samples digested under the same conditions. Each batch 5 consisted of 10 samples, a reagent blank and a secondary reference material (a homogenised potato material prepared in house). The reagent blank was used to check for contamination from digestion vessels, whereas the secondary reference material was used to check the efficiency of the digestion to secure uniformity of the sample matrix from batch to batch. The validation of the analytical method is described in Bibak et al. (1999). 0

Results and discussion

The concentrations of 58 elements in potatoes from the conventionally and organically cultivated plots are shown in Table 4.

25

The selection of data, data processing and statistical analysis were performed as described in Example 1.

The score plot of a discriminant partial least square regression (DPLS) of mean values of 30 14 elements is shown in Figure 5. Elements only participating as noise in the regression are excluded.

TABLE 4 (continued)

Potatoes (Solan urn tuberosum, Folva)

Conventionally produced (μg/kg) Organically produced (μg/kg)

Element P- n Mean Median Sdev Confin Mean Median Sdev Confivalue dence dence (%) interval interval

Cs133 9664 44 b>c b>c 31 091 45 14 b>c 48 14

Ba138 2011 70 43 40 13 29 71 44 43 14 32

La 139 145^* 68 011 0093 0068 0016 69 006 0053 0036 00084

Ce140 8256 47 34 12 18 5 55 51 b>c 13 35

Pr141 584 71 0037 0021 0053 0012 72 0033 0028 0041 00094

Nd146 220* 69 009 0075 0072 0017 72 0057 0053 0053 0012

Sm147 3964 71 0055 0043 0049 0011 72 0034 0023 0045 001

Eu153 1973 59 023 018 024 006 57 02 015 024 0061

Gd 157 1775 71 0033 0017 0052 0012 59 0029 0032 0033 00085

Tb159 4698 71 00095 00077 0013 00029 72 00079 00066 001 00023

Dy 163 7593 69 002 00098 0053 0012 67 00093 b>c 0035 00085

Ho 165 8296 72 b>c 000013 00086 0002 72 b>c b>c 00076 00018

Er166 2992 72 b>c b>c 0035 0008 72 b>c b>c 0036 00083

Tm169 6806 72 000055 b>c 00093 00021 72 b>c b>c 00037000085

Yb172 7878 71 001 b>c 0015 00036 71 00098 b>c 0015 00035

Lu 175 1709 72 00027 000023 0012 00028 71 00023 b>c 00088 00021

Hf 178 4417 63 035 024 051 013 67 026 02 027 0064

Ta181 5113 55 39 25 41 11 58 21 15 19 048

Re 185 6759 72 0034 0018 0064 0015 69 00069 b>c 0026 00062

Ir 193 3778 71 0027 00079 0081 0019 71 00035 00012 0027 00063

Pt195 2728 68 04 026 054 013 69 0096 0082 016 0039

Au 197 7711 60 12 081 098 025 72 072 06 057 013

TI205 159^* 59 1 097 063 016 71 057 054 02 0046

Bι209 4093 67 022 0077 044 011 70 0086 0049 014 0033

Th232 43 62 066 045 087 022 70 023 012 027 0064

U238 2929 61 00084 b>c 016 0041 65 b>c b>c 0045 0011 It is clearly seen from the score plot in Figure 5, that the potatoes are separated in two distinct groups: conventionally and organically produced potatoes. The potatoes are marked with second letter "E" when organically cultivated, and marked with second letter "C" when conventionally cultivated. The 14 elements used in order to separate the pota- toes are shown in the loading plot in figure 6.

The experiment confirms that the described procedure makes it possible to separate organically produced potatoes from conventionally produced potatoes that have been cultivated on the same geographical location.

REFERENCES

Bibak A., Stϋrup S., Haahr V., Gundersen P., and Gundersen V. Concentrations of 50 Major and Trace Elements in Danish Agricultural Crops Measured by Inductively Coupled Plasma Mass Spectrometry. 3. Potato (Solanum tubersum Folva). J. Agric. Food Chem. (1999) 47:7, 2678-2684.

Box, G. E. P., Hunter, W. G. and Hunter, J. S. Statistics for experimenters. John Wiley and Sons Inc., New York, 1978.

Brown, S. D., Bear, S. R. Jr, and Blank, T. B. Chemometrics Anal. Chem. 64, (1992) 22R-49R.

Coomans, D. and D.L. Massart, Analytica Chimica Acta, 138, (1982), 15.

Coomans, D., D.L. Massart and L. Kaufman, Analytica Chimica Acta, 112, (1979), 97.

Haykin, Neural Networks, Macmillan, New York, 1994.

Klockenkaemper, R., von Bohlen, A., and Moens, L. X-Ray Spectrom. 29, 119-129 (2000).

Lachenbruch, P.A. Discriminant analysis, Haffner Press, New York, 1975.

Malinowski, E. R. Factor Analysis in Chemistry, 2nd ed., John Wiley & Sons 1991.

Manly, B.F.J. Multivariate Statistical Methods : A Primer, Chapman & Hall, London, 1994.

Martens, H. and Naes, T. Multivariate calibration, John Wiley & Sons, New York, 1989.

Massart, D. L. and Kaufmann, L. Interpretation of analytical data by the use of cluster analysis, John Wiley & Sons, New York, 1983.

Massart, D. L., Vandeginste, B. G. M., Deming, S. N., Michotte, Y. and Kaufman, L. Che- mometrics: a textbook, Elsevier, Amsterdam, 1988. Mel nger, M Correspondance analysis The method and its application Chemometrics and Intelligent Laboratory Systems, 2, (1987), 61-77

Montaser, Akbar George Washington University, Washington, D C 20052, USA, Inductively Coupled Plasma Mass Spectrometry, Wiley-VCH, Inc , 1998, Chap 1 and 6

Piggott, J R Statistical procedures in food research, Elsevier Applied Science, 1986 Sharaf, M A , lllman, D L and Kowalski, B R Chemometrics, John Wiley & Sons, New York, 1986

Sjostrom, M , S Wold, and B Soderstrom PLS discriminant plots In Pattern recognition in practice II E S Gelsema and L N Kanal, Eds , Elsevier, Amsterdam, 1986, p 486

Skoog, Douglas A , Donald M West and F James Holler, Fundamentals of Analytical Chemistry, sixth edition, Saunders College Publιshιng1992, Chap 24

Wold, S , Esbensen, K , and Geladi, P Principal component analysis Chemometrics Intel! Lab Syst 2, (1987), 37

Claims

1. A method for determining that a product has been organically produced, said method comprising the steps of

(a) collecting a sample of the assumed organically produced product,

(b) subjecting said sample to an elemental analysis to obtain data for the elemental content, and

(c) establishing the fact that the product has been organically produced by subjecting said data to a data analysis and showing that the data differs from elemental content data obtained from a conventionally produced product, and/or

(d) establishing the fact that the product has been organically produced by subjecting said data to a data analysis and showing that the data are similar to elemental content data obtained from an organically produced product.

2. A method according to claim 1 wherein the product is selected from the group consist- ing of a horticultural product, an agricultural product and an aquacultural product.

3. A method according to claim 2 wherein the horticultural product is a vegetable.

4. A method according to claim 3 wherein the vegetable is selected from the group con- sisting of onion, pea, cabbage, sprout, potato, carrot, and salad.

5. A method according to claim 2 wherein the horticultural product is a fruit.

6. A method according to claim 2 wherein the horticultural product is a flower.

7. A method according to claim 2 wherein the agricultural product is of animal origin.

8. A method according to claim 7 wherein the agricultural product is selected from the group consisting of egg, milk, milk products, meat, meat products, a whole animal and a part of an animal.

9. A method according to claim 2 wherein the product is an agricultural product of vegetable origin.

10. A method according to claim 9 wherein the product is selected from the group con- 5 sisting of barely, wheat, rye and oats.

1 1. A method according to claim 2 wherein the aquacultural product is selected from the group consisting of fish, shellfish and crustaceans.

10 12. A method according to claim 1 wherein the elemental analysis is performed by applying a mass spectrometer.

13. A method according to claim 12 wherein the mass spectrometer is selected from the group consisting of a high resolution inductively coupled plasma mass spectrometer (HR- 15 ICPMS), a quadrupole inductively coupled plasma mass spectrometer (Q-ICPMS), an ion cyclotron resonance mass spectrometer, an ion trap mass spectrometer, a time-of-flight mass spectrometer, a direct reader mass spectrometer, a multiple-collector magnetic mass analyser, a twin-quadrupole instrument, and a collision cell interface.

20 14. A method according to claim 1 wherein the elemental analysis is performed by applying an optical spectrometer.

15. A method according to claim 14 wherein the optical spectrometer is selected from the group consisting of an inductively coupled plasma optical emission spectrometer (ICP-

25 OES, ICP-AES), an inductively coupled plasma atomic fluorescence spectrometer (ICP- AFS), a flame atomic absorption spectrometer (AAS), a flame atomic emission spectrometer (AES), an electrothermal atomisation atomic absorption spectrometer (GFA), a X-ray fluorescence spectrometer (XRF), and a total reflection X-ray fluorescence spectrometer (TXRF).

30

16. A method according to claim 1 wherein the data analysis is performed by a manual selection.

17. A method according to claim 1 wherein the data analysis is performed by statistical 35 analysis.

18. A method according to claim 17 wherein the data analysis is a multivariate analysis.

19. A method according to claim 18 wherein the multivariate analysis is principal component analysis (PCA).

20. A method according to claim 18 wherein the multivariate analysis is partial least square regression including discriminant partial least squares regression (DPLS).

21. A method according to claim 18 wherein the multivariate analysis is selected from the group consisting of Analysis of variance (ANOVA), Linear discriminant analysis (LDA),

Soft independent modelling of class analogy (SIMCA), K-nearest neighbours (KNN), Cluster analysis (CA), Canonical Variates Analysis (CVA), Artificial neural networks (ANNs), Factor analysis, Multidimensional scaling (MDS), and Correspondence factor analysis (CFA).