Simultaneous detection of Ponceat 4R and tartrazine in food using adsorptive stripping voltammetry on an acetylene black nanoparticle-modified electrode

BACKGROUND: Ponceau 4R and tartrazine have been widely used in foodstuffs. However, they are pathogenic if they are excessively consumed. Therefore, the detection of Ponceat 4R and tartrazine is quite important. RESULTS: A sensitive and rapid electrochemical method was developed for the simultaneous detection of Ponceat 4R and tartrazine using anodic adsorptive stripping voltammetry and based on the strong enhancement effect of acetylene black nanoparticle. For Ponceat 4R, the linear range was from 0.05 to 4 mg kg^?1, and the limit of detection was 0.03 mg kg^?1. For tartrazine, the linear range was from 0.15 to 18 mg kg^?1, and the limit of detection was 0.1 mg kg^?1. The relative standard deviation was 3.8% and 4.7% for 10 successive measurements of 1 mg kg^?1 Ponceau 4R and tartrazine. The method was used to determine Ponceat 4R and tartrazine in soft drinks, and recovery was in the range of 92.4–104.8%. CONCLUSION: At the acetylene black nanoparticle‐modified electrode, the oxidation current signal of Ponceau 4R and tartrazine greatly increase. This new method is sensitive, rapid, simple and feasible. Copyright © 2011 Society of Chemical Industry     

Estimated intake of intense sweeteners from non-alcoholic beverages in Denmark, 2005

In 2005 International Standards Organization. ISO 17025. 2005. General requirements for the competence of testing and calibration laboratories, Copenhagen: Danish Standard [Google Scholar], 76 out of 177 analysed samples of non-alcoholic beverages were found to contain the intense sweeteners cyclamate, acesulfame-K, aspartame, and saccharin. The content of cyclamate did not exceed the now permitted maximum level in the European Union of 250 mg l^?1 in soft drinks. The estimated intake of the sweeteners was calculated using the Danish Dietary Survey based on 3098 persons aged 1–80 years. The estimated intake with 90th percentiles of 0.7, 0.8 and 0.2 mg kg^?1 body weight day^?1 for acesulfame-K, aspartame, and saccharin, respectively, was much lower than the acceptable daily intake values of 15, 40, 7, and 2.5 mg kg^?1 body weight day^?1 for acesulfame-K, aspartame, and saccharin, respectively, and on the same level as in the similar investigation from 1999. In contrast to the 1999 investigation, the 90th percentile of the estimated cyclamate intake in 1–3 year olds with 3.7 mg kg^?1 body weight day^?1 was in 2005 lower than the acceptable daily intake of 7 mg kg^?1 body weight day^?1. However, the 99th percentile for 1–3 year olds with 7.4 mg kg^?1 body weight day^?1 still exceeded the acceptable daily intake slightly. The 90th percentile for the whole population with 0.9 mg kg^?1 body weight day^?1 was halved compared with 1999. The reduction in the European Union of the maximum permitted level for cyclamate from 400 to 250 mg l^?1 has brought the intake of cyclamate in small children down to well below the acceptable daily intake value.     

Estimate of intake of benzoic acid in the Belgian adult population

An exposure assessment was performed to estimate average daily benzoic acid intake for Belgian adults. Food consumption data were retrieved from the national food-consumption survey. As a first step, individual food-consumption data were multiplied with the maximum permitted use levels for benzoic acid per food group (Tier 2). As a second step, a label survey to identify the foods where benzoic acid is effectively used as an additive and a literature review of the possible occurrence of benzoic acid as a natural substance were performed. With this information, a refined list of foods was drafted for the quantification of benzoic acid, which was performed by a high-performance liquid chromatography (HPLC) method, optimized and validated for this purpose. Individual food-consumption data were then multiplied with the actual average concentrations of benzoic acid per food group (Tier 3). Usual intakes were calculated using the Nusser method. The mean benzoic acid intake was 1.58 mg kg^?1 body weight day^?1 (Tier 2) and 1.25 mg kg^?1 body weight day^?1 (Tier 3). In Tier 2, men exceeded the acceptable daily intake (ADI) of 5 mg kg^?1 body weight day^?1 at the 99th percentile. The greatest contributors to the benzoic acid intake were soft drinks. Benzoic acid as a natural substance represents only a small percentage of the total intake. The results show that actual benzoic acid intake is very likely to be below the ADI. However, there is a need to collect national food-consumption data for children as they might be more vulnerable to an excessive intake.     

Estimated intake of the sweeteners,acesulfame-K and aspartame,from soft drinks,soft drinks based on mineral waters and nectars for a group of Portuguese teenage students

In a survey of levels of acesulfame-K and aspartame in soft drinks and in light nectars, the intake of these intense sweeteners was estimated for a group of teenage students. Acesulfame-K was detected in 72% of the soft drinks, with a mean concentration of 72 mg l^?1 and aspartame was found in 92% of the samples with a mean concentration of 89 mg l^?1. When data on the content of these sweeteners in soft drinks were analysed according to flavour, cola drinks had the highest mean levels for both sweeteners with 98 and 103 mg l^?1 for acesulfame-K and aspartame, respectively. For soft drinks based on mineral water, aspartame was found in 62% of the samples, with a mean concentration of 82 mg l^?1 and acesulfame-K was found in 77%, with a mean level of 48 mg l^?1. All samples of nectars contained acesulfame-K, with a mean concentration of 128 mg l^?1 and aspartame was detected in 80% of the samples with a mean concentration of 73 mg l^?1. A frequency questionnaire, designed to identify adolescents having high consumption of these drinks, was completed by a randomly selected sample of teenagers (n = 65) living in the city of Coimbra, in 2007. The estimated daily intakes (EDI) of acesulfame-K and aspartame for the average consumer were below the acceptable daily intakes (ADIs). For acesulfame-K, the EDI was 0.7 mg kg^?1 bw day^?1 for soft drinks, 0.2 mg kg^?1 bw day^?1 for soft drinks based on mineral waters, and 0.5 mg kg^?1 bw day^?1 for nectars, representing 8.0%, 2.2%, and 5.8% of the ADI, respectively. A similar situation was observed for aspartame. In this way, the EDI for soft drinks was 1.1 mg kg^?1 day^?1, representing only 2.9% of the ADI. In respect of nectars, the EDI was 0.2 mg kg^?1 bw day^?1, representing 0.5% of the ADI. Soft drinks based on mineral waters showed the lowest EDI values of 0.3 mg kg^?1 bw day^?1, accounting for 0.7% of the ADI.     

Determination of the caffeine contents of various food items within the Austrian market and validation of a caffeine assessment tool (CAT)

The caffeine content of 124 products, including coffee, coffee-based beverages, energy drinks, tea, colas, yoghurt and chocolate, were determined using RP-HPLC with UV detection after solid-phase extraction. Highest concentrations of caffeine were found for coffee prepared from pads (755?mg?l^?1) and regular filtered coffee (659?mg?l^?1). The total caffeine content of coffee and chocolate-based beverages was between 15?mg?l^?1 in chocolate milk and 448?mg?l^?1 in canned ice coffee. For energy drinks the caffeine content varied in a range from 266 to 340?mg?l^?1. Caffeine concentrations in tea and ice teas were between 13 and 183?mg?l^?1. Coffee-flavoured yoghurts ranged from 33 to 48?mg?kg^?1. The caffeine concentration in chocolate and chocolate bars was between 17?mg?kg^?1 in whole milk chocolate and 551?mg?kg^?1 in a chocolate with coffee filling. A caffeine assessment tool was developed and validated by a 3-day dietary record (r ^2?=?0.817, p?<?0.01) using these analytical data and caffeine saliva concentrations (r ^2?=?0.427, p?<?0.01).     

Occurrence and infant exposure assessment of nitrates in baby foods marketed in the region of Lisbon,Portugal

Commercial baby food labelled as from organic or conventional origin, including vegetable-based baby foods, fruit purees and fruit juices (n?=?80), were analysed for nitrate content by an in-house validated HPLC method. Nitrate contents ranged from 5 to 230?mg?kg^?1 with a mean concentration of 102?mg?kg^?1 for vegetable-based baby foods, and a median of 5?mg?kg^?1 for both fruit purees and juices. One sample of vegetable-based baby food was higher than the legislated value (200?mg?kg^?1). There were no significant differences between average nitrate levels in analysed samples regarding both farming systems. The estimated nitrate intake through baby foods for a mean nitrate concentration of 47?mg?kg^?1 ranged between 0.5 (15% of ADI) and 1.3?mg?kg^?1?bw?day^?1 (35% of ADI). The ADI level was exceeded (107–146% of ADI) only for the 95th and 99th percentiles of nitrate concentration.     

Aflatoxin M1 in Spanish infant formulae: occurrence and dietary intake regarding type,protein-base and physical state

The incidence of aflatoxin M₁ (AFM₁) in 69 different infant formulae marketed in Spain between 2007 and 2008 was studied and dietary intake estimated. Samples were analysed using an HPLC method coupled with fluorescence detection after immunoaffinity column clean-up. The toxin was detected in 26 formulae (37.7%) at levels below the permissible limit set by EC legislation, giving a range of 0.6–11.6 ng kg^?1 with a mean value of 3.1 ng kg^?1. Increasing occurrence was found in those formulae produced by the less complex manufacturing processes affecting casein/whey protein ratio: pre-term, 14.3%; starter, 35.3%; follow-up, 42.1%; toddler 87.6%; while hypoallergenic and lactose-free were totally exempt. Additionally, the influence of main protein source and physical state (powdered and ready-to-use formula) on AFM₁ occurrence was evaluated leading to similar conclusions. Dietary AFM₁ weekly intake was observed to be stable around 1 ng kg^?1 bw for standard formula and 0.1 ng kg^?1 bw for pre-term feeding.     

Occurrence of patulin and 5-hydroxymethylfurfural in apple sour,which is a traditional product of Kastamonu,Turkey

Apple sour is a traditional product of Kastamonu, Turkey. It is consumed by spreading on bread or drinking after diluting with water. The aim of this study was to determine patulin (PAT) and 5-hydroxymethylfurfural (HMF) in apple sour. This study is the first to evaluate the occurrence of PAT and HMF in apple sour. The samples were extracted with ethyl acetate using liquid-liquid extraction technique. PAT and HMF were determined by HPLC with UV detection. PAT was detected in all samples, and the PAT level in 94.9% of samples was found to be equal or greater than the legal limit for juice concentrates. The mean value for PAT was found to be 284 ± 307 μg kg^?1. PAT levels in 13 of 39 samples were in the range of 100 ≤ x < 200 μg kg^?1, two samples were in the range of 0 ≤ x < 50 μg kg ?1 and two samples were in the range of 1000 ≤ x < 1500 μg kg¹. HMF levels of all samples were above the legal limit for solid molasses. The mean value for HMF was found to be 16215 ± 13317 mg kg^?1. HMF levels of 10 of 39 samples were determined to be in the range of 10000 ≤ x < 20000 mg kg^?1, eight samples were in the range of 20000 ≤ x < 30000 mg kg^?1 and only three samples were in the range of 100 ≤ x < 1000 mg kg^?1. There was a significant and inverse relationship between HMF and pH of the samples. These results indicate that consumption of apple sour is a considerable risk in terms of HMF and PAT toxicity.     

顶空-气相色谱法快速测定饮用酒类中甜蜜素含量

该实验建立了顶空-气相色谱（HS-GC）快速检测饮用酒类中甜蜜素的方法。样品经加热去除大部分乙醇和挥发性物质,样品提取液经衍生后,用顶空-气相色谱仪分析,以甜蜜素衍生产物环己烯为目标物,外标法定量。结果表明,当甜蜜素在饮用酒类中对应含量为1～100 mg/kg时线性关系良好（R2=0.998 9）,检出限和定量限分别为0.1 mg/kg和0.3 mg/kg,加标回收率为94.3%～107.1%,相对标准偏差（RSD）为2.5%～7.1%(n=6),精密度试验结果RSD为3.1%～4.9%。该方法简单、高效、环境友好、可靠、准确,可用于各种饮用酒类中甜蜜素的快速定量检测。     

Sensory properties of some synthetic high‐intensity sweeteners in water solutions

BACKGROUND: When sucrose is eliminated from products to meet consumers' demand for calorie‐free products and substitutes are used, sweetness becomes an important characteristic. The objective of this study was to compare sensory properties of four sucrose substitutes: aspartame, acesulfame K, sodium saccharin and sodium cyclamate in water solutions. RESULTS: The lowest concentrations of sweeteners such as 0.35 g kg^?1 for aspartame and acesulfame K showed an equi‐sweet level relative to approx. 55 g kg^?1 aqueous sucrose solution, whereas 2.3 g kg^?1 sodium cyclamate and 0.4 g kg^?1 sodium saccharin relative to approx. 65 g kg^?1. Aspartame had almost the same sweetness potency as acesulfame K, whereas the remaining sweeteners differed completely. The four sweeteners showed their specific sensory profiles. Besides the characteristics of sweetness and bitterness, metallic and astringent attributes plus warming and cooling effects were found. CONCLUSIONS: Of the four sweeteners, aspartame and sodium cyclamate can be considered as the best sucrose substitutes due to their similar sensory profiles to the sucrose. The data showing the interdependence between sweetness intensity of the sweeteners and sucrose are useful as a quick and easy indicator of the sweetener amount having equi‐sweet levels relative to sucrose. Copyright © 2009 Society of Chemical Industry     

A survey of styrene monomer levels in foods and plastic packaging by coupled mass spectrometry—automatic headspace gas chromatography

An extensive UK survey of styrene monomer levels in styrene based plastic packaging materials and their contained foods (133 samples) has been carried out, examining a wide range of retail foods of different brand names and including yogurts, creams, salads, coleslaws, soft cheeses, margarines, hot and cold beverages from dispensing machines, spreads, fresh and cooked meats, candied fruits, fresh strawberries, and take-away ‘fast’ foods. Analysis by headspace gas chromatography of styrene levels in the plastic containers showed levels of monomer ranging from 16 to 1300 mg kg^?1 although the majority of containers (73%) had styrene levels in the range 100–500 mg kg^?1 and only five plastic tubs had levels exceeding 1000 mg kg^?1. Analysis of the food contents of the plastic containers by automated headspace gas chromatography with single ion monitoring mass spectrometric detection showed levels of the monomer ranging from < 1 μg kg^?1 to 200 μg kg^?1, although the majority of foods (77%) had styrene levels below 10 μg kg^?1, and 26% of the total number analysed had levels below 1 μg kg^?1.     

Sodium benzoate and potassium sorbate preservatives in food stuffs in Iran

A high-performance liquid chromatography method was applied for the determination of the levels of benzoate and sorbate in 400 food samples, including pickled cucumbers, canned tomato pastes, sour cherry jams, soft drinks, fruit juices and dairy products (UF-Feta cheeses, Lighvan cheeses, lactic cheeses, yogurts and doogh). The results showed that 270 (67.5%) of all samples contained benzoate ranging from 11.9 to 288.5 mg kg^?1 in lactic cheese and fruit juice, respectively. The levels of sorbate in 98 (24.5%) of the samples were 20.1 to 284.3 mg kg^?1 in doogh and fruit juice, respectively. Moreover, benzoate was detected in all dairy products ranging from 11.9 mg kg^?1 in lactic cheese to 91.2 mg kg^?1 in UF-Feta cheese. A low concentration of benzoate could originate naturally, due to specific biochemical mechanisms during cheese, yogurt and doogh maturation. In conclusion, a minimum level for benzoate in dairy products should be defined in the legislation.     

Fluoride content of some common South Indian foods and their contribution to fluorosis

Fluoride contents of 98 food items commonly used in South India were determined with special reference to the feeding habits of remote rural populations in 41 villages of Anantapur District, Andhra Pradesh, India, where dental and skeletal fluorosis is endemic. Fluoride up to 4.5 mg kg^?1 was found in irrigation and drinking water supplies. Thirty-two locally grown food items had generally higher fluoride contents (ranging jiom 0.2 to 11.0 mg kg^?1) with the notable exception of coconut water where even traces of fluoride could not be detected. The combined daily intake of fluoride jiom food and drinking water in the local population was found to range jiom 2.2 mg to 7-3 mg (0.05–0.32 mg kg^?1 BW). The role played by food composition, cooking habits of the local population, general poverty and illiteracy in contributing to the prevalence of fluorosis in schoolchildren in the age group 6–18 years (even in areas with drinking water supplies within the permissible levels as per WHO standards) is discussed.     

Boron concentrations in selected foods from borate‐producing regions in Turkey

As part of a large‐scale programme to investigate the health effects of exposure to borate minerals in Turkey, boron concentrations in vegetables, fruits and some other foods were determined. From all borate‐producing regions of the country, 22 species of fruit, 17 species of vegetable, 12 species of cereal, legume and oilseed, three species of herbs and six types of other food were collected in 1999 and analysed for boron concentrations using the Azomethine H method. The results were compared with 37 types of food collected from areas well distant from the borate‐producing areas. The total number of samples evaluated was 420. Pistachio (67.0 mg kg^?1) had the highest concentration of boron, followed by grape leaf (60.48 mg kg^?1), sour cherry (57.03 mg kg^?1), quince (38.78 mg kg^?1), peach (34.49 mg kg^?1), grape (20.70 mg kg^?1), green beans (19.49 mg kg^?1), unripe peach (18.92 mg kg^?1) and parsley (10.24 mg kg^?1) from the borate‐producing regions. A large number of foods contained boron at higher concentrations than those reported in the literature, most probably due to the high boron content of these Turkish soils. According to the results of t‐tests, the area itself does not have a significant effect (p > 0.05) on boron concentration in all foods evaluated so far. Copyright © 2003 Society of Chemical Industry     

Dietary exposure of secondary school students in Hong Kong to benzoic acid in prepackaged non-alcoholic beverages

This study evaluated the dietary exposure of secondary school students in Hong Kong to benzoic acid from pre-packaged non-alcoholic beverages. Exposure was estimated using local food consumption data of secondary school students obtained by a semi-quantitative food frequency questionnaire in 2000 and the benzoic acid level detected in pre-packaged beverages, including soft drink (both diet/light and regular types), fruit juice, soy milk, Chinese tea and coffee/tea) available locally in late 2006. The estimated dietary exposure to benzoic acid from pre-packaged beverages of average and high consumers (95^th percentile) was 0.31 and 0.97 mg kg^?1 bw day^?1, respectively. These exposures accounted for 6.1 and 19.3% of the acceptable daily intake (ADI: 0–5 mg kg^?1 bw) of benzoic acid for average and high consumers, respectively. As in other countries, soft drinks contributed most to dietary exposure to benzoic acid from pre-packaged beverages in Hong Kong.     

Rapid quantitative method for total brominated vegetable oil in soft drinks using ion chromatography

A simple, quantitative and rapid method for total brominated vegetable oil (BVO) using ion chromatography (IC) with suppressed conductivity detection was developed and successfully applied to soft drinks with results expressed as inorganic bromide anion. The procedure involves extraction of BVO with diethyl ether and treatment with zinc dust in a solution of acetic acid, giving recoveries ranging between 92.5 and 98.5%. The calibration curves obtained were linear with correlation coefficients (r ²) of 0.998, a coefficient of variation (CV) of less than 5% and limit of detection (LOD) and limit of quantification (LOQ) of 250 and 750?µg?l^?1, respectively. The method was successfully applied to the determination of BVO in several commercial soft drinks which were found to contain BVO in the range 1.8–14.510?mg?l^?1. The method has less sources of error compared to previously published methods.     

Validation of the procedure for the determination of maduramicin in concentrates,premixes, and feeds by liquid chromatography

A single laboratory validation was carried out for the determination of maduramicin in concentrates, premixes, and feed. The method comprised sample extraction of maduramicin, derivatization with dansylhydrazine and liquid chromatography with ultraviolet light detection. The limit of detection (LOD) and limit of quantification (LOQ) were 0.4 and 1.0 mg kg^?1, respectively. The repeatability expressed as the average difference between the results of duplicate measurements was 5.9% at the concentration level of 1% (concentrate), 7.1% at the concentration level of 1 g kg^?1 (premix), and 11% with the feed containing maduramicin with the nominal concentration of 5 mg kg^?1 and feed spiked at the concentration level of 1 mg kg^?1. The relative standard deviations for the within-laboratory reproducibility (RSD_W) were 9.2%, 16%, 18%, and 17% at the concentration levels of 1%, 1 g kg^?1, 5 mg kg^?1, and 1 mg kg^?1, respectively. The measurement uncertainties were ±0.2%, ±0.3 g kg^?1, ±1.9 mg kg^?1, and ±0.3 mg kg^?1 at the same concentration levels, respectively.     

Sensitive determination of cadmium in brown rice and spinach by flame atomic absorption spectrometry with solid-phase extraction

A sensitive flame atomic absorption spectrometry (FAAS) method was developed for the determination of cadmium (Cd) in brown rice and spinach. The method involves extraction with 1?M hydrochloric acid (HCl), followed by a selective pre-concentration by solid-phase extraction (SPE). The pH of the loading sample solution was adjusted to 4.0 for the brown rice and to 5.0 for the spinach. The masking agents, tartrate and citrate, were required for the spinach before pH adjustment. The SPE step achieved a 20-fold enrichment of the sample solution. The limits of quantification (LOQs) were 0.0054?mg?kg^?1 for the brown rice and 0.0022?mg?kg^?1 for the spinach, being more sensitive than those of AOAC Official method 999.10. A single-laboratory validation was performed by testing spiked samples at 0.04 and 0.08?mg?kg^?1 for the brown rice, and 0.02 and 0.04?mg?kg^?1 for the spinach. The average recoveries were 93.3–96.9% with relative standard deviations (RSDs) of 4.1–8.2% for brown rice, and 90.5–91.9% with RSDs of 5.8–10.0% for spinach.     

Influence of branch bending on sugar,organic acid and phenolic content in fruits of ‘Williams’ pears (Pyrus communis L.)

Selected sugars, organic acids and phenolic compounds were analysed in mature fruits of ‘Williams’ pears using high‐performance liquid chromatography. Fruits were harvested from the branches of trees tested in three treatments: branches were bent in summer 2003 (1 September), in spring 2004 (15 May) and control (branches were not bent). Pears contained up to 73.54 g kg^?1 fresh weight (FW) of fructose, 9.42 g kg^?1 FW of glucose, 7.94 g kg^?1 FW of sucrose and 24.59 g kg^?1 FW of sorbitol. Major organic acids were (in order of descending quantity) citric, malic, shikimic and fumaric acid (up to 3.05 g kg^?1 FW, 2.24 g kg^?1 FW, 71.79 mg kg^?1 FW and 0.49 mg kg^?1 FW, respectively). Chlorogenic acid (280.86–357.34 mg kg^?1 FW) was the predominant phenolic acid, followed in concentration (mg kg^?1 FW) by syringic acid (95.46–131.32), epicatechin (46.55–83.09), catechin (25.67–44.81), vanillic acid (1.87–3.48), sinapic acid (0.83–1.72) and caffeic acid (0.72–1.04). Significant differences in content of fructose, sorbitol, total sugars, catechin, epicatechin, sinapic acid, syringic acid, and a sum of determined phenolic compounds were observed among the treatments. Fruits from summer bending branches had the lowest content of individual sugars, citric acid and phenolic compounds and the highest content of malic, shikimic and fumaric acid. The highest content of fructose, sorbitol, sucrose, total sugars, caffeic acid, catechin, epicatechin and syringic acid were determined in the fruits from the spring treatment. In the control treatment the highest content of glucose, citric acid, chlorogenic acid, sinapic acid, vanillic acid, as well a sum of determined phenolics, were observed. The lowest content of fumaric acid was in the spring treatment and of malic and shikimic acid in the control. Copyright © 2006 Society of Chemical Industry     

Development of a Headspace-Solid Phase Microextraction-Gas Chromatography/Mass Spectrometry (HS-SPME-GC/MS) Method for the Determination of Benzene in Soft Drinks

A method based on headspace-solid phase microextraction and gas chromatography with mass spectrometry has been developed and validated for the determination of benzene in soft drinks. The extraction step was optimized using a rotatable central composite design including the following experimental variables: extraction temperature, extraction time, sample weight, and salt concentration. The optimized procedure, which was carried out at 30 °C during 30 min by using a 75 μm carboxen-polydimethylsiloxane fiber, showed good linearity within the concentration range 0–25 μg?kg^?1 (r ² ?>?0.999), mean recoveries from 97.5 to 103.1 %, and coefficients of variation from 1.5 to 13.4 % for repeatability and from 1.5 to 15.7 % for within-laboratory reproducibility. Limits of detection and quantification were calculated at 0.02 and 0.08 μg?kg^?1, respectively. The method was applied to determine the concentrations of benzene in 77 samples of beverages from the Brazilian market. Levels from <0.08 to 10.84 μg?kg^?1 were obtained, which are comparable to those verified in other countries. Most of the samples (72.2 %) contained benzene up to 1 μg?kg^?1.