Determination of niacin in infant formula and wheat flour by anion-exchange liquid chromatography with solid-phase extraction cleanup.

Niacin content must be included on food labels of infant formula products and bakery products containing enriched flour. Liquid chromatographic (LC) determination of niacin in complex food matrixes is complicated by the presence of endogenous compounds that absorb at the commonly used wave-length of 260 nm. Also, the presence of particulate matter in the standard sulfuric acid extraction procedure results in reduced life of LC columns and precolumns. A simple, rapid, solid-phase extraction (SPE) procedure for separation and cleanup of niacin from a complex food matrix digest has been developed. By using a vacuum manifold with the SPE column system, multiple samples can be processed quickly and efficiently for LC analysis, compared with gravimetric column cleanup. Sulfuric acid sample digest is passed over an aromatic sulfonic acid cation-exchange (ArSCX-SPE) or a sulfonated Florisil SPE column. Niacin is eluted with 0.25M sodium acetate-acetic acid, pH 5.6 buffer in vacuo. LC chromatograms of the resulting eluate are free of interference from other components absorbing at 260 nm at the retention time of niacin. Validation of the method was obtained from agreement of analytical results on available reference materials. For both SPE methods, values for niacin in SRM 1846 Infant Formula (milk-based powder) were within uncertainty ranges of the certified value. Use of several calibration procedures (the LC computer program, a peak area response graphic standard curve, or the method of standard additions) with both SPE procedures resulted in niacin values for 3 RM-Wheat Flours (not certified for niacin) in agreement (90-105%) with their respective values reported in the literature. Several commercial wheat flours showed a broad 260 nm interference, resulting in high niacin values. Niacin recoveries from spiked soy-based liquid infant formulas ranged from 95-107% with the ArSCX-SPE column. Calibration curves of niacin were linear up to 400 micrograms/mL, with a detection limit of 0.2 microgram/mL.

Vitamin C content of foods: sample variability.

A recent survey of foods that constitute the major sources of vitamin C in the American diet yielded information on the total content of this vitamin as well as the amount of its two forms, ascorbic acid and dehydroascorbic acid (DHAA) in these foods. Samples of individual foods showed a surprising large range of vitamin content even for foods collected from the same regions of the country and from the same source. The amount of DHAA in the different foods varied from approximately 10% to 20% of the total vitamin content. The large range of values for the vitamin content in a given food suggests further that in human-diet studies, when the major sources of vitamin C are from a few foods, daily analyses are required for the necessary precision.

Liquid chromatographic determination of amprolium in poultry feed and premixes using postcolumn chemistry with fluorometric detection.

Two extraction and liquid chromatographic procedures are presented which separate amprolium from compounds in poultry feed or premixes that could interfere with its fluorometric determination. The procedures are based on earlier work on the determination of thiamine in food samples. Amprolium is extracted from feed with a hexane-aqueous sulfosalicylic acid mix, separated on a C18 column, and detected fluorometrically after postcolumn derivatization. For premixes, water extraction is used. Values for the amprolium content of poultry feed obtained with these procedures are in good agreement with those obtained with AOAC official methods. It is suggested that these methods with suitable modifications may be of use for routine analysis of amprolium in feeds. The overall methods are rapid and appear to give reasonable results.

Application and flexibility of robotics in automating extraction methods for food samples.

Laboratory robotic technology has made it possible to automate the manually intensive operations associated with the extraction of vitamins from food. The modular approach to robotics allows the conversion from one extraction procedure to another by a simple addition or replacement of a module plus reprogramming. This is illustrated for the extraction of vitamins C and B1 from food samples. Because many of the organic micronutrients are unstable, storage and extraction conditions must be established to stabilize labile compounds if the full capabilities of robotics are to be realized.

HPLC analysis with fluorometric detection of vitamin C in food samples.

A high performance liquid chromatographic (HPLC) procedure has been developed for the analysis of ascorbic acid and dehydroascorbic acid in complex matrices. Separation is accomplished with an anion-exchange resin and fluorescent detection is achieved through post-column inline chemistry, involving oxidation of ascorbic acid to dehydroascorbic acid followed by reaction with o-phenylenediamine to form a fluorescent product. Lower limits of detection for both forms of vitamin C are well below the levels found in the usual food sources of this vitamin. The extraction procedures developed yield clean samples for analysis with minimal loss of the vitamers during the analytical procedures. Recoveries are in the range of 90-107%. The results obtained with this HPLC procedure agree well with those obtained with a modified version of the classical procedure of Deutsch and Weeks. A variety of foods including fruit juices, vegetables, and fruits were analyzed.

Liquid chromatographic determination of vitamin B-6 in foods.

Chromatographic analysis for vitamin B-6 in complex samples imposes certain requirements on the analyst, who must extract completely the bound, unstable vitamers without loss, remove interfering compounds, and provide clean extracts for analysis. The analyst also has to contend with the problems inherent in all methods, such as sample collection, storage, preparation, and homogenization. However, chromatography provides a means of identifying and quantitating all forms of the vitamin, and thus provides the possibility of addressing the problem of the bioavailability of specific vitamers. It also allows automation, which is absolutely essential in coping with the large numbers of samples that are generated in areas such as quality control. These factors are all addressed here, and chromatographic results for various meat and other food products are presented to illustrate the variations in vitamin content that occur from sample to sample, the agreement with microbiological results, and that liquid chromatography (LC) has come of age in dealing with complex biological samples, such as food and food products.

Forms of vitamin B6 in human milk.

A previously developed high performance chromatographic method has been modified slightly and used to determine the forms of vitamin B6 in human milk. The chromatographic traces are free of compounds that would interfere with the two principal forms found, pyridoxal and pyridoxal phosphate. The method is fast and reliable; it yields recoveries of from 90 to 106% for the vitamers and agrees with results obtained on the same samples with the standard microbiological assay.

B6 vitamer analysis in human plasma by high performance liquid chromatography: a preliminary report.

An extraction procedure that can be used successfully with a high performance liquid chromatography sample clean-up and analytical method is outlined for the determination of B6 vitamer and pyridoxic acid concentrations in human plasma. The method is a modification of ones developed for food and animal tissue investigations. Chromatographic traces, relatively clean of interfering compounds are obtained. The method has the advantage of speed and reproducibility and yields recoveries of from 90 to 100% for all vitamers and pyridoxic acid.

Laminar dispersion in flow-injection analysis.

Simple expressions are given for the dispersion and the travel times of samples in simple flow-injection analysis systems. The sum of these two quantities is the total residence time of the sample in the system. The expressions are based on numerical solutions of the diffusion-convection equation. Preliminary experiments are in agreement with the derived simple expressions, as are peak curve shapes. Diffusion coefficients can be obtained in a straightforward manner.