Search

Background: Ponceau 4R (E124) and carmoisine (CMS; E122) are frequently utilized azo synthetic dyes in the food industry owing to their aesthetically pleasing coloration and broad consumer acceptability. It is imperative to prioritize environmentally favorable technologies for quantifying these dyes, as excessive consumption of these poses significant health risks., Objective: The primary objective of this research was to establish a reversed-phase (RP)-HPLC method that could simultaneously detect Ponceau 4R and CMS, implementing green analytical chemistry (GAC) and analytical quality by design (AQbD), using an ultrasound-assisted extraction (UAE) technique in commercial food samples., Methods: An Agilent Eclipse Plus column (C18, 250 × 4.6 mm id, 5 µm) was utilized for effective separation with a mobile phase of ethanol-acetate buffer pH 5 (60:40, v/v), flow rate of 1 mL/min, and detection wavelength of 515 nm. Critical variables selected for method optimization were ethanol percentage and flow rate, determined using central composite design (CCD). In order to adhere to the 12 principles of green chemistry, hazardous solvents were substituted with ethanol, which is distinguished by its ease of use, effectiveness, and ecological sustainability. The greenness assessment was conducted utilizing the green analytical procedure index (GAPI), analytical eco-scale (AES), and analytical greenness metrics (AGREE)., Results: The respective retention times for Ponceau 4R and CMS were 2.276 and 3.450 min. The recovery rate of Ponceau 4R and CMS fluctuated between 70% and 102% and 80% and 102%, respectively, across various marketed food samples. The procedure passed validation in accordance with the International Conference on Harmonization Q14 guidelines., Conclusion: The devised method demonstrates that the validation parameters like linearity, precision, sensitivity, and reproducibility are within the specified limits of ICH guidelines. The greenness assesment tools GAPI, AES, and AGREE produced the most favorable results., Highlights: In future, environmentally sustainable, solvent-based, robust AQbD methodologies for assessing varieties of food colorants may be adopted and improved commercially., (© The Author(s) 2024. Published by Oxford University Press on behalf of AOAC INTERNATIONAL. All rights reserved. For permissions, please email: journals.permissions@oup.com.)

Background: Brilliant Blue and Sunset Yellow, two highly water-soluble synthetic food dyes, are the most popular food dyes used and consumed. Although they are not highly toxic, some health problems can be observed when excessive amounts of food products containing these dyes are consumed. Objectives: The aim of the study was to develop a simultaneous UV-Vis combined solid-phase extraction method, based on the adsorption onto Amberlite XAD-8 resin, for determination of Brilliant Blue and Sunset Yellow dyes. Methods: Sample solution was poured into the reservoir of the column and permitted to gravitationally pass through the column at 2 mL/min flow rate. Adsorbed dyes were eluted to 5 mL of final volume with 1 mol/L HNO₃ in ethanol solution by applying a 2 mL/min flow rate. Dye concentrations of the solution were determined at 483 and 630 nm for Sunset Yellow and Brilliant Blue, respectively. Results: The detection limits of the method for Brilliant Blue and Sunset Yellow were determined as 0.13 and 0.66 ng/mL, respectively. Preconcentration factor was 80. Brilliant Blue contents of real food samples were found to be between 11 and 240 μg/g. Sunset Yellow concentrations of foodstuffs were determined to be between 19 and 331 μg/g. Conclusions: Economical, effective, and simple simultaneous determination of Brilliant Blue and Sunset Yellow was achieved by using a solid-phase extraction combined UV-Vis spectrometry method. Highlights: The method is applicable and suitable for routine analysis in quality control laboratories without the need for expert personnel and high operational costs because the instrumentation is simple and inexpensive.

Edible colorants, as an important part of food additives, can not only enhance the sensorial attributes of foodstuffs, but can also increase one's appetite. They hold a very important position in food processing. In the past decades, the illegal addition of the inedible colorants has become one of the major issues of food safety. Industrial dyes, especially some azo dyes, are illegal additives frequently found in foodstuffs. They cannot provide any nutrients for the human body and even have toxicity, carcinogenic, or mutagenic effects, which may cause serious damage to consumers' health. There is an increasing demand to detect the inedible azo dyes in foodstuffs for health and safety reasons. In this review, the types and the physicochemical properties of the inedible azo dyes adulterated in foodstuffs and beverages are summarized, and the emphases are focused on the sample pretreatment methods and analytical techniques for monitoring of these dyes in foodstuffs and beverages.

A method for simutaneously detecting 8 oil-soluble and 10 water-soluble (3 basic and 7 acidic) illegal dyes in foods was developed. The sample was mixed with water, followed by methanol and tetrahydrofuran. Transesterification with sodium methoxide was applied to the mixture, which allowed the triglycerides in the sample to be converted to fatty acid methyl esters. This treatment resulted in a biphasic mixture. Oil-soluble dyes and fatty acid methyl esters were deposited in the upper organic phase, which was cleaned using a silica-gel solid-phase extraction (SPE) column to remove the fatty acid methyl esters from the solution. The water-soluble dyes were deposited in the aqueous phase, and an Oasis hydrophilic-lipophilic-balanced SPE column was used to remove polar matrix components from the solution. The resulting dyes were subsequently analyzed via LC-diode-array detection using a single method. The practical LODs of the samples were defined as the lowest spiked dye concentrations at which the similarity coefficient for the spectra of the LC test solution and the corresponding reference standard solution were greater than 0.99, thus affording LODs of 0.5-1.0 µg/g. Recoveries of the dyes at a spiking level of 5.0 µg/g from soft drink, chili sauce, and mustard were generally greater than 70%. Recoveries from paprika powder were between 33 and 103%.

An ultra-performance LC (UPLC) method was developed to determine the manufacturing intermediates and subsidiary colors in the monosulfo monoazo color additives D&C Red No. 6 and D&C Red No. 7 and their lakes. This method is intended for use in batch certification of the color additives by the U. S. Food and Drug Administration to ensure that each lot meets published specifications for coloring drugs and cosmetics. The intermediates are 2-amino-5-methylbenzenesulfonic acid (PTMS) and 3-hydroxy-2-naphthalenecarboxylic acid (3-hydroxy-2-naphthoic acid). The subsidiary colors are 3-hydroxy-4-[(4-methylphenyl)azo]-2-naphthalenecarboxylic acid (unsulfonated subsidiary color) and 1-[(4-methylphenyl) azo]-2-naphthalenol (4-methyl Sudan I). The analytes were identified by comparing their UPLC retention times and UV-Vis absorption spectra with those of standards. Validation studies showed that calibration curves were linear (average R2=0.9994), and recoveries were 96-106%. Average LOD was 0.0014-0.0061% and average LOQ was 0.0047-0.020%. Results for RSD at the specification levels ranged from 0.67 to 5.79%. Survey analyses of 42 samples from 14 domestic and foreign manufacturers yielded results by the new UPLC method and a previously reported HPLC method that were consistent within experimental error. The new UPLC method provided increased sensitivity, faster analysis times, and improved separations compared to the HPLC method.

Two novel spectrophotometric determination procedures based on retention of Allura Red onto Amberlite XAD-1180 and XAD-16 resins for its preconcentration, purification, and separation were developed. Analytical parameters of the methods including pH, eluent type, sample volume, and sample and eluent flow rates, were investigated and optimized. Interference effects of some cations, anions, and widely used food dyes were also investigated. Detection limits of the two methods were found to be 1.2 and 1.5 μg/L for XAD-1180 and XAD-16 columns, respectively, under optimum conditions. Linear calibration curve ranges of the methods were 0.4-8.0 and 0.5-6.0 μg/mL of Allura Red for XAD-1180 and XAD-16 resins, respectively. Preconcentration factors were found as 80 for both the XAD-1180 and XAD-16 columns using maximum sample volume and minimum eluent volume. RSDs of the methods were below 6% throughout all experiments. All absorbance measurements were performed at 506 nm. Validations of the methods were performed comparatively with determination of the Allura Red contents of some foodstuff, pharmaceutical, and energy drink samples. Allura Red concentrations in investigated solid and liquid samples ranged from 298 to 501 μg/g and 53.8 to 508 μg/mL, respectively. Satisfactory results were obtained from the real samples analysis. Allura Red contents of samples were determined to be highly similar using the two extraction methods. Comparisons of the methods were performed by analysis of Allura Red contents of the real samples. In addition to analytical parameters, adsorption isotherm studies were performed for the two kinds of Amberlite resins. It was observed that developed methods fit the linear form of the Freundlich adsorption isotherm model. All of the experimental results suggested that the developed SPE procedures are suitable for separation, preconcentration, and determination of Allura Red in solid and liquid matrixes.

A sensitive electrochemical method was developed for the simultaneous determination of Brilliant Blue (BB) and tartrazine (Tz) using an ionic liquid-modified expanded graphite paste electrode (IL-EGPE). The IL-EGPE was prepared by mixing ionic liquid-expanded graphite composite (IL-EG) with solid paraffin. Compared with the EGPE, the IL-EGPE remarkably enhanced the electrocatalytic oxidation signals of BB and Tz. Under optimal experimental conditions, the designed IL-EGPE exhibited wide linear responses to BB and Tz ranging from 5.0×10(-9) to 4.0×10(-6) M and 1.0×10(-8) to 1.0×10(-6) M, respectively. The detection limits for BB and Tz were 2.0×10(-9) M (1.6 ng/mL) and 3.3×10(-9) M (1.8 ng/mL) at an S/N of 3, respectively. This electrode showed good reproducibility, stability, and reusability. The proposed method was successfully applied in the simultaneous determination of BB and Tz in a soft drink with satisfactory results.

Color additives are dyes, pigments, or other substances that can impart color when added or applied to foods, drugs, cosmetics, medical devices, or the human body. These substances must be pre-approved by the U.S. Food and Drug Administration (FDA) and listed in the Code of Federal Regulations before they may be used in FDA-regulated products. Both domestic and imported cosmetic products sold in interstate commerce fall under FDA jurisdiction, and FDA's district laboratories use a combination of analytical methods for identifying or confirming the presence of potentially violative color additives. We have developed a qualitative method for identifying 29 water- and methanol-soluble color additives in various types of cosmetic products. The color additives are extracted with combinations of methylene chloride, methanol, acetic acid, and water and are identified by LC with photodiode array detection. Estimated LOD values ranged from 0.1 to 1.5 mg/L. A survey of lip products, nail polishes, eye products, blushes, body glitter, face paints, bath products, creams, and toothpastes identified permitted and non-permitted color additives. Our new LC method is intended to supplement the visible spectrophotometry and TLC methods currently used by FDA's district laboratories and will help optimize the use of time, labor, and solvents.

The present method utilizes a simple pretreatment step, cleanup on polyamide SPE cartridges, and HPLC resolution on reversed-phase C18 for the detection of the three basic nonpermitted dyes encountered in food matrixes. Polyamide cartridges were chosen because both acidic and basic dyes can be cleaned up due to their amphoteric nature. Analysis was performed on a reversed-phase C18 micro-Bondapak column using the isocratic mixture of acetonitrile-sodium acetate with a flow rate of 1.5 mL/min and a programmable lambda(max) specific visible detection to monitor colors, achieving higher sensitivity and expanded scope to test multicolor blends. All the colors showed linearity with the regression coefficient, from 0.9983 to 0.9995. The LOD and LOQ ranged between 0.107 and 0.754 mg/L and 0.371 and 2.27 mg/L or mg/kg, respectively. The intraday and interday precision gave good RSDs, and percentage recoveries in different food matrixes ranged from 75 to 96.5%. The study demonstrates that the use of a combination of a simple SPE cleanup and HPLC resolution with UV-Vis end point detection was successful in screening the presence of these three basic nonpermitted dyes individually or in blend, in a variety of food matrixes.

A cloud point extraction method is reported using LC/MS for the determination of regulated water-soluble food colors (Allura Red, Sunset Yellow, erythrosine, and tartrazine) and banned fat-soluble synthetic azo dyes (Sudan I, II, III, and IV; Red B; 7B; Black B; Red G; Metanil Yellow; and Rhodamine B). The extraction of all 14 colors was carried out with cloud point extraction using the nonionic surfactant Triton X 114. Optimized conditions for cloud point extraction were 3% Triton X 114 (w/v), 0.1 M ammonium acetate, and heating at 50 degrees C for 30 min. This approach proved effective in giving quantitative recoveries from a diverse range of food matrixes, and optimized LC gave baseline chromatographic separation for all colors including Sudan IV and Red B. Single-laboratory validation was performed with spiking into liquid matrixes (wine and homemade wine), semiliquid matrixes (sauce and homemade paprika paste), and solid matrixes (spice and homemade chili powder) using the respective blank matrixes for matrix-matched calibration. The LOQ values for water-soluble colors were in the range of 15-150 mg/kg, and for the fat-soluble colors, 0.1-1.5 mg/kg. The mean recovery values were in the range of 69.6-116.0% (except Allura Red and Sunset Yellow in wine, for which recoveries were lower). The mean RSDs for colors were in the range of 4.0-14.8%. A small survey was conducted of samples of confectionery products, dried fruits, wines, bitter sodas, juices, sauces, pastes, and spices, which demonstrated the applicability of the method to a diverse selection of real food samples. Allura Red was detected in strawberry jelly and Sunset Yellow in artificial saffron.

A simple and sensitive HPLC method has been developed for the simultaneous determination of eight permitted food colors and five commonly encountered nonpermitted colors in various food commodities, including sugar-, fat-, and starch-based food matrixes. The method uses a specific food category-based cleanup/treatment procedure before color extraction to avoid the interference of food matrixes, and to obtain the optimal color extraction. Analysis was performed on a reversed-phase C18 micro-Bondapak column with ammonium acetate and acetonitrile gradient elution as the mobile phase; a programmable lamda max-specific visible detection was used to monitor colors to obtain the higher sensitivity and expanded scope needed for multicolor blends having diverse absorption maxima. All colors showed good linearity, with regression coefficients of 0.9974-0.9999. The LOD and LOQ values ranged from 0.01 to 0.12 mg/L, and from 0.04 to 0.83 mg/L or mg/kg, respectively. The intraday and interday precision tests produced good RSD values, and the recoveries from different food matrixes ranged from 82 to 104%. The method offers high sensitivity for analysis of a wide variety of food matrixes containing a broad scope of multicolor blends. Two nonpermitted colors, orange II and metanil yellow, were found. Also, a number of samples contained permitted colors at levels two- to seven-fold higher than those prescribed.

Rapid determination of Sudan I in foodstuffs is very important because it was found to be a carcinogen and its use in the food industry was banned. A rapid, sensitive, and convenient electrochemical method was developed for the determination of Sudan I based on the distinctive properties of mesoporous SiO2. The electrochemical responses of Sudan I were investigated. A sensitive oxidation peak was observed for Sudan I, and the peak currents greatly increased at the mesoporous SiO2-modified electrode, which can be attributed to its large surface area and high accumulation efficiency. The effects of pH, amount of mesoporous SiO2, scan rate, accumulation potential, and time were examined on the oxidation signals of Sudan I. The linear range was from 4.00 x 10(-8) to 2.00 x 10(-5) M, and the LOD was 7.50 x 10(-9) M. The newly developed method was successfully used to detect and quantify Sudan I in hot chili powder and juice samples.

A reversed-phase liquid chromatographic (LC) method was developed to determine 4 impurities in the monosulfo monoazo color additives D&C Red No. 6, D&C Red No. 7, and their lakes. Manufacturing intermediates determined are 2-amino-5-methylbenzenesulfonic acid (PTMS) and 3-hydroxy-2-naphthalenecarboxylic acid (3-hydroxy-2-naphthoic acid). Subsidiary colors determined are the unsulfonated and uncarboxylated, unsulfonated homologs of the dye (the unsulfonated subsidiary color and 4-methyl-Sudan I). The new method uses lithium oxalate as part of a novel system to dissolve the water-insoluble color additives. The 4 impurities were identified by LC and UV-visible spectrophotometry; the 4-methyl-Sudan I was confirmed by LC/mass spectrometry. LC peak area calibrations were generally linear (R > 0.999), and recoveries ranged from 90 to 105%. The limits of determination used were 0.02% each for PTMS and 3-hydroxy-2-naphthoic acid, 0.1% for the unsulfonated subsidiary color, and 0.002% for 4-methyl-Sudan I. The relative standard deviations at the specification levels ranged from 0.7 to 22.7%. Survey analyses of 26 samples of certified D&C Red Nos. 6 and 7 straight colors and lakes from 11 domestic and foreign manufacturers yielded results that agree with those previously obtained using 3 other methods.

A rapid planar chromatographic method for identification and quantification of 25 water-soluble dyes in food was developed. In a horizontal developing chamber, the chromatographic separation on silica gel 60F254 high-performance thin-layer chromatography plates took 12 min for 40 runs in parallel, using 8 mL ethyl acetate-methanol-water-acetic acid (65 + 23 + 11 + 1, v/v/v/v) mobile phase up to a migration distance of 50 mm. However, the total analysis time, inclusive of application and evaluation, took 60 min for 40 runs. Thus, the overall time/run can be calculated as 1.5 min with a solvent consumption of 200 microL. A sample throughput of 1000 runs/8 h day can be reached by switching between the working stations (application, development, and evaluation) in a 20 min interval, which triples the analysis throughput. Densitometry was performed by absorption measurement using the multiwavelength scan mode in the UV and visible ranges. Repeatabilities [relative standard deviation (RSD), 4 determinations] at the first or second calibration level showed precisions of mostly < or = 2.7%, ranging between 0.2 and 5.2%. Correlation coefficient values (R > or = 0.9987) and RSD values (< or = 4.2%) of the calibration curves were highly satisfactory using classical quantification. However, digital evaluation of the plate image was also used for quantification, which resulted in RSD values of the calibration curves of mostly < or = 3.0%, except for two < or = 6.0%. The method was applied for the analysis of some energy drinks and bakery ink formulations, directly applied after dilution. By recording of absorbance spectra in the visible range, the identities of the dyes found in the samples were ascertained by comparison with the respective standard bands (correlation coefficients > or = 0.9996). If necessary for confirmation, online mass spectra were recorded within a minute.

A method using simple extraction and 2-directional high-performance thin-layer chromatography (HPTLC) was developed for the simultaneous determination of curcumin, metanil yellow, and sudan dyes in turmeric, chili, and various mixed curry powder formulations. The method offers resolution (Rf) of turmeric pigments, namely, curcumin (0.77), demethoxycurcumin (0.69), bis(demethoxy)curcumin (0.61), and the synthetic dye metanil yellow (0.05) by the first-directional mobile phase, chloroform-methanol (9 + 1, v/v). The resolution (Rf) of sudan I (0.30) and sudan IV (0.23) was achieved by the second-directional mobile phase, toluene-hexane-acetic acid (50 + 50 + 1, v/v/v). Natural pigments of both turmeric and chili showed no interference in the detection and quantification of synthetic colors. The limit of detection and limit of quantification values for curcumin, metanil yellow, sudan I, and sudan IV were 17.39, 42.90, 15.45, and 7.01 and 52.71,130.0, 46.80, and 21.24 ng/spot, respectively. Analysis of a few market samples showed the presence of metanil yellow (1.5-4.6 mg/g), sudan I (4.8-12.1 mg/g), and sudan IV (0.9-2.0 mg/g) in loose turmeric and chili samples, whereas the curcumin content in turmeric and mixed curry powder samples ranged from 6.5 to 36.4 and from 0.3 to 1.9 mg/g, respectively. The method is relatively simple, offers reasonable sensitivity, and can be used to screen a large number of samples.

Square wave (SW) voltammetry was used to explore the adsorption properties of the food additive dye Allura Red on a hanging mercury drop electrode (HMDE). By using the adsorptive stripping voltammetric approach, we developed a sensitive electroanalytical method for the determination of this azo dye. A well-developed voltammetric peak probably related to the cathodic reduction of the azo moiety was obtained in pH 9 Britton-Robinson (B-R) buffer at 613 mV. Cyclic voltammetric studies indicated that the reduction process was irreversible and primarily controlled by adsorption. The adsorptive voltammetric signal was evaluated with respect to various experimental conditions; the optimized values were supporting electrolyte, B-R buffer; pH 11; accumulation time, 180 s; accumulation potential, 0.0 V; scan rate, 900 mV/s; pulse amplitude, 75 mV; and SW frequency, 90 Hz. Adsorptive voltammetric peak current showed a linear response for Allura Red in the concentration range of 2.5 x 10(-8) to 2.0 x 10(-7) mol/L (r = 0.998). The limit of detection was 8.5 x 10(-9) mol/L (4.2 ng/mL), the precision in terms of relative standard deviation was 1.3%, and the mean recovery was 102%. Possible interferences by several substances usually present in food products such as food additive azo dyes (E110, E102), gelatin, natural and artificial sweeteners, preservatives, and antioxidants were also evaluated. The proposed electrochemical procedure was successfully applied to the determination of this food dye in commercially available candy and a soft drink. The results were compared by statistical evaluation with those obtained by a reference spectrophotometric method.

This collaborative study was conducted to determine the total monomeric anthocyanin concentration by the pH differential method, which is a rapid and simple spectrophotometric method based on the anthocyanin structural transformation that occurs with a change in pH (colored at pH 1.0 and colorless at pH 4.5). Eleven collaborators representing commercial laboratories, academic institutions, and government laboratories participated. Seven Youden pair materials representing fruit juices, beverages, natural colorants, and wines were tested. The repeatability relative standard deviation (RSDr) varied from 1.06 to 4.16%. The reproducibility relative standard deviation (RSDR) ranged from 2.69 to 10.12%. The HorRat values were < or = 1.33 for all materials. The Study Director recommends that the method be adopted Official First Action.

The differences in color coordinates obtained from the use of different spectral features in the calculation are studied. Seven groups of food (olive oil, vanilla milkshake, brandy, honey, grape juice, vinegar, and orange juice) and an anthocyanin petunidin solution (natural pigment present in several vegetables) with different pH values were selected. Tristimulus values were calculated by considering the different sources of errors (truncation, abridgement, or different bandwidths). Results obtained were corrected by using the methods recommended (see References). These methods of abridgment work well in general, although in some circumstances (20 nm bandwidth or larger), specific spectral weighting functions have to be used to obtain a negligible error. Therefore, it is interesting to know how much difference can be expected from those factors in order to avoid confusion between color differences attributable to instruments and those attributable to actual color changes.

A simple method has been developed for the extraction, separation, and determination of synthetic colors in ice cream samples. The process involves the breakdown of emulsion by neutral detergents (Triton X-100 and Tween 20) followed by extraction with petroleum ether for removal of fat. The aqueous colored solution obtained is treated with 5% acetic acid, and the uptake of color is carried out by a wool-dyeing technique. The color is eluted from the wool with 5% ammonia solution, the solution is evaporated to dryness, and the residue is dissolved in 60% ethanol for paper chromatography using trisodium citrate-ammonia-water (2 + 5 + 95, w/v/v) as the mobile phase. The colored spots from the paper chromatogram are cut and eluted with 60% ethanol, and the absorbance is measured at the respective lambda maximum corresponding to the Rf value of the appropriate standard. The recoveries of 6 colors, including sunset yellow FCF (SSYFCF), tartrazine, carmoisine, ponceau 4R, brilliant blue FCF (BBFCF), and fast green FCF from spiked samples with either detergent were found to be >90%. However, recoveries of erythrosine were 21 and 65% with Triton X-100 and Tween 20, respectively. Indigo carmine could not be recovered at all because of its fugitive property in 5% ammonia solution, which is used to strip the color from the wool. The sensitivity of the method with the use of Tween 20 is 1 ppm (1 microg/g) for the colors in spiked ice cream samples. With this method, we analyzed samples of 20 branded colored ice cream. The results showed the presence of tartrazine (8.4-43.3 ppm), SSYFCF (23.5-117.6 ppm), carmoisine (traces-53.2 ppm), erythrosine (3.5 ppm), and BBFCF (4.1 ppm) in the ice cream samples. Apart from 2 samples of tuttifruity, all of the ice cream samples showed the presence of permitted synthetic colors below the permissible level of 100 ppm established by the Prevention of Food Adulteration Act of India.

A new methodology was developed for determination of caramel in spirits aged in oak casks. The method is based on differences between the electronic spectra of oak aqueous alcoholic extracts and caramel solutions in the same solvent. The data were treated by 2 different approaches: the simplest one was based on the plot of caramel concentration versus the ratio of absorbance at 210 and 282 nm; the other was based on a partial least squares (PLS) calibration model using the first derivative of the spectral data. Both methodologies were applied to analysis of 159 aged spirit samples. The mean caramel content of several Brazilian sugar cane spirits (cachaça) and all United States whiskies was smaller than that of Scottish whiskies and other brandies from several countries. Correlation was good between caramel concentrations for the same sample calculated by the 2 methods. The uncertainties following PLS and the absorbance ratio method were 0.01 and 0.03 g/L, respectively, for a sample containing 0.45 g/L caramel. Treatment of UV-VIS spectra by pattern recognition using hierarchical clustering analysis and principal components analysis allowed discrimination of the samples as a function of their caramel content. It was possible to distinguish U.S. whiskies from other whiskies, but a clear differentiation among Brazilian cachaças as a function of their geographic origin was not feasible. Small caramel quantities as low as 0.08 g/L were clearly detected by these methodologies.

The preparations of digested samples of certifiable color additives by dry ashing and wet digestion for arsenic analysis by hydride generation atomic absorption spectrometry (AAS) were compared. The dry ashing technique was based on the preparation used in ASTM D4606-86 for determination of As and Se in coal. The acid digestion method used nitric and sulfuric acids heated by microwaves in sealed vessels. The digested color additives were analyzed for As by using hydride generated from sodium borohydride mixed with the acidified solution on a flow injection system leading to an atomic absorption spectrometer. Dry ashing was preferable to wet digestion because wet digestion yielded poor recoveries of added As. Dry ashing followed by hydride generation AAS gave determination limits of 0.5 ppm As in the color additives. At a specification level of 3 ppm As, the precision of the method using dry ashing was +/- 0.4 ppm (95% confidence interval).

A multiresidue method to analyze liquid whole milk for 59 compounds was developed. The method involves a single extraction and cleanup strategy for many classes of compounds--organophosphorus compounds (OPs), organochlorine compounds (OCs), N-methylcarbamates (MCs), etc. Initial extraction is performed with ethanol-ethyl acetate as solvent and sodium sulfate as drying agent. A portion of the extract is concentrated to an oily consistency, and target analytes are partitioned into acetonitrile. Further cleanup is achieved by sequential solid-phase extractions--octadecyl (C18)-bonded silica cartridges followed by aminopropyl (NH2)-bonded silica cartridges. The solvent is exchanged to acetone for analysis by gas chromatography (GC) with electrolytic conductivity, flame photometric (FPD) or mass spectrometric (MS) detection and to methanol for analysis by liquid chromatography with postcolumn derivatization. Typical limits of detection (LODs; peak-to-peak signal-to-noise ratio > or = 3) were 0.3 ppb for OPs, 0.9 ppb for OCs and MCs, and 9 ppb for mass-selective detection. From the level of quantitation (LOQ = 3.33 x LOD) to 10 x LOQ, linear instrument response was observed for all detectors except the FPD which required a second-order calibration curve. Average recoveries for spikes at the LOQ ranged from 69 to 127% with standard deviations of about 10%. Similar accuracy and precision were observed for fortifications at 5 x LOQ and 10 x LOQ. The method was used to analyze 20 milk samples from various liquid milk processing plants. Incurred residues were confirmed by high-resolution GC/MS and GC/MS/MS.

The Food Analysis Performance Assessment Scheme (FAPAS) organized by a Secretariat of the UK Ministry of Agriculture, Fisheries, and Food has checked the proficiency of analytical laboratories for foods and feeds from 1990 to 1996. FAPAS was started for UK laboratories but was expanded worldwide at the request of analysts in other countries who did not have a home-based scheme. Thirteen thousand homogeneity-checked test materials were issued, covering a very wide range of analytes, including pesticides, toxins, veterinary drug residues, trace and nutritional elements, food colors, preservatives, sweeteners, alcohol congeners, fatty acids, nitrate, and proximate analysis. Participants returned 85% of requested data, and 47,000 z-score proficiency assessments were made, of which 81% were satisfactory. Evidence is presented of improvements in overall analytical ability with increased participation in proficiency testing in the areas of proximate analysis; organochlorine pesticide analysis; and lead, mercury, and acesulfame-K analyses. Little improvement was shown in other analytical areas such as calcium analysis. Overall accuracies for analysis of specific pesticides and specific trace elements in the circulated test materials were compared.

A competitive enzyme immunoassay (EIA) for carminic acid was investigated. Monoclonal anticarminic acid antibody was obtained from A/J mice immunized with carminic acid-human immunoglobulin G (IgG) conjugate. Carminic acid was extracted with distilled water from beverage, jelly, candy, pasta sauce, yogurt, or ice cream samples. Ham or fish paste samples were digested with pronase, then carminic acid was extracted from samples with sodium hydroxide solution. The extract was diluted more than 10-fold with 1% gelatin in borate buffer solution. Microtiter plates were coated with carminic acid-bovine serum albumin (BSA) conjugate or just BSA. Goat anti-mouse IgG(H+L)-peroxidase complex was used as a second antibody, and 3,3',5,5'-tetramethylbenzidine was used as a substrate for the peroxidase. The working range for quantitative analysis was 0.3-10 ng/mL, and the detection limit was 0.2 micrograms/g original sample. Recoveries of carminic acid by this assay were > 95% for milk beverage and jelly, and > 85% for yogurt and fish paste. Carminic acid was detected in 7 of 26 red-colored commercial food products and ranged from 3.5 to 356 micrograms/g. This EIA system also responded to the structural analogue of carminic acid, laccaic acid.

A method is described for the determination of the intermediates and a related impurity in D&C Red No. 36 by reversed-phase liquid chromatography. This method may be used to ensure that limits set forth in the Code of Federal Regulations on the amounts of these 3 impurities in the color are not exceeded. The pigment is dissolved in boiling dioxane and then precipitated. The filtrate is chromatographed by isocratic elution, and then the column is washed and reequilibrated. Impurities were identified as 2-chloro-4-nitroaniline (2-Cl-4-NA), 2-naphthol, and 2,4-dinitroaniline (2,4-DNA) by comparison of their retention times and spectra with those of standards. Peak area calibrations were linear to at least 0.375% 2-Cl-4-NA, 1.25% 2-naphthol, and 0.025% 2,4-DNA, all with zero intercepts. At the specification levels, 99% confidence limits were 0.30 +/- 0.006% for 2-Cl-4-NA, 1.0 +/- 0.03% for 2-naphthol, and 0.020 +/- 0.0004% for 2,4-DNA. The limits of determination calculated from calibration data were 0.019% for 2-Cl-4-NA, 0.10% for 2-naphthol, and 0.0014% for 2,4-DNA at the 99% confidence level. Recoveries were 100-104% for 2-Cl-4-NA added to purified D&C Red No. 36, 100% for 2-naphthol, and 100-110% for 2,4-DNA; relative standard deviations were 0.8-3.4%. A survey of certified D&C Red No. 36 samples showed that the batches contained higher levels of intermediates than were determined previously by a cellulose column method in which the pigment was not dissolved.