C985 35 PDF [PDF]

Zitiervorschau

FOOD COMPOSITION AND ADDITIVES

Extension of Dry Ash Atomic Absorption and Spectrophotometric Methods to Determination of Minerals and Phosphorus in Soy-Based, Whey-Based, and Enteral Formulae (Modification of AOAC Official Methods 985.35 and 986.24): Collaborative Study COOK: JOURNAL OF AOAC INTERNATIONAL VOL. 80, NO. 4, 1997 KATHLEEN K. COOK U.S. Food and Drug Administration, Office of Food Labeling, Division of Science and Applied Technology, 200 C St, SW, Washington, DC 20204 Collaborators: R. Allen; A. Choudhry; R. Fleener; F.E. Greene; C. Johnson; N. Miller-Ihli; R. Powell; W.-L. Yip

Eight laboratories participated in a collaborative study of AOAC Official Method 985.35, Minerals in Ready-to-Feed Milk-Based Infant Formula and Pet Foods, Atomic Absorption Spectrophotometric Method; and 7 laboratories participated in a study of AOAC Official Method 986.24, Phosphorus in Milk-Based Infant Formula, Spectrophotometric Method, to extend these methods to infant formulae (other than milk-based) and enteral products. Three ready-to-feed soy-based formulae and 2 soybased powder formulae were chosen to represent the plant matrix. A whey-based formula and a casein-based enteral formula were also included in the study. Soy formulae containing nearly identical concentrations of particular elements were matched, and an application of the Youden “closely matched pair” approach was used to estimate repeatability parameters. Average reproducibility values were as follows: calcium, 9.3% %; copper, 9.7% %; iron, 5.5% %; potassium, 4.0% %; magnesium, 5.2% %; manganese, 10.6% %; sodium, 4.7% %; phosphorus, 10.5% %; and zinc, 7.3% %. At similar analyte concentrations, the betweenlaboratory variabilities compared well with those reported for the official methods. Most repeatability and reproducibility parameters compared well with the original collaborative study. AOAC Official Methods 985.35 and 986.24 have been modified to extend their applicability to infant formulae (other than milk-based) and enteral products.

Submitted for publication October 28, 1996. The recommendation was approved by the Methods Committee on Food Nutrition, and was adopted by the Official Methods Board of the Association. See “Official Methods Board Actions” (1997) J. AOAC Int. 80, 35A, and “Official Methods Board Actions” (1997) Inside Laboratory Management, March issue.

collaborative study was performed to confirm that the dry ash atomic absorption method for determination of Ca, Mg, Zn, Cu, Fe, Mn, K, and Na in ready-to-feed milk-based infant formula, AOAC Official Method 985.35, and the spectrophotometric method for phosphorus in milkbased infant formula, AOAC Official Method 986.24, could be extended to determination of these minerals in all types of infant formula and enteral products.

A

Collaborative Study In the original collaborative studies of methods of analysis for infant formula, 3 liquid ready-to-feed (RTF) milk-based formulae were analyzed and the same analysis was repeated on a different day (1, 2). For the study reported here, 3 RTF formulae and 2 soy-based powders were selected to represent a plant source matrix, which was not included in the original study. A whey-based powder and a casein-based enteral formula were also included to represent other matrixes derived from a milk base. Eight collaborators, including the author, used Method 985.35 (3), and 7 collaborators, including the author, used Method 986.24 (4) to analyze soy, whey, and enteral formulae. In addition, each collaborator was asked to add a spiking solution to one of the following: the enteral formula, an RTF soy-based formula, a soy powder formula, or the whey powder formula. The protocol of the International Union of Pure and Applied Chemistry (IUPAC; 5) was used to interpret study results. This protocol uses sequential application of the Cochran and Grubb’s tests to the data to identify outliers. The same protocol was used to determine recovery. Results (Tables 1–11) show that Methods 985.35 and 986.24 are reliable and applicable to other formulae.

Powders The contents of 6 cans from the same lot of a particular brand of powder formula were composited by adding half of their 12 to 16 oz. contents to a large beaker. After thorough

mixing, these composites were apportioned into screw-top plastic jars for shipment to collaborators. For the replicate portion to be spiked, 6 g of either soy powder or whey powder was weighed into a screw-top plastic jar. The material was to be added to the ashing vessel with the spiking solution. The minimum weight specified for use for powders was 1.5 g.

985.35 Minerals in Infant Formula, Enteral Products, and Pet Foods, Atomic Absorption Spectrophotometric Method Final Action 1985 Final Action 1988 Revised First Action 1997

Liquids Collaborators analyzed 4 liquid formulae: 8 fl. oz. cans and bottles of the same lot, with labels removed, representing 2 of the RTF soy products and the single enteral formula, and a 32 fl. oz. can with the label printed on the can representing the third RTF soy formula. No attempt was made to eliminate can-to-can variation within lots because of the instability of the product once opened. Collaborators analyzing a spiked liquid were given directions to weigh a 30 g portion of the 32 fl. oz. RTF soy formula or the enteral formula into the ashing vessel and to add the accompanying spiking solution. The only formula identified for the collaborators was the 32 fl. oz. can with the label printed directly on the can.

Spiking Solution A spiking solution containing all 9 minerals in a plastic screw-top vial was included. Collaborators were directed to add this solution to the designated formula to be spiked and then to rinse the vial 3 times, collecting the rinses in the ashing vessel. Because P was included in the same spiking solution and only one portion was to be spiked, collaborators were required to use the same preparation step for P and the other minerals. An in-house study was conducted to investigate whether results for phosphorus would be compromised by this procedure. No differences were found between results obtained from using the extracts from the atomic absorption method for minerals and the results of the official phosphorus method in the inhouse study. The results are summarized in Table 12.

Within-Laboratory Variability Liquid RTF soy formulae and soy powders with nearly identical concentrations of particular elements were used to estimate within-laboratory variability by mechanical application of the Youden “closely matched pair” approach.

Blank Determination To identify problems due to contamination, collaborators were asked to run 2 blanks through each method.

Concentration Ranges Estimates of concentration ranges for all minerals in the products were supplied to each collaborator. However, actual concentrations found for some elements were outside these ranges. As a consequence, one collaborator reported 2 determinations for elements found to be outside the ranges supplied. The first reported values were used in this report.

(Applicable to Ca, Mg, Fe, Zn, Cu, Mn, Na, and K.) Caution: See Appendix B: safety notes on safe handling of acids. Dispose of waste solvents in an appropriate manner compatible with applicable environmental rules and regulations. Method Performance: See Table 985.35A for method performance data.

A. Principle Organic matrix is destroyed by dry ashing in muffle furnace. Remaining ash is dissolved in diluted acid and analyte is determined by atomic absorption spectrophotometry (AAS).

B. Apparatus (a) Glassware.—Thoroughly clean all glassware by soaking overnight in 20% HNO3. Rinse all glassware 3× with distilled-deionized or 18 MΩ resistance H2O. (b) Evaporation dish.—100 mL unetched Vycor (or Pt), flat-bottom, with pour spout; capable of withstanding temperatures up to 600°C. (c) Atomic absorption spectrophotometer.—Equipment should be well maintained with good response per unit concentration, for example, 0.200 abs or above 4 mg/L Cu. (d) Furnace.—With pyrometer to control temperature range of 250°–600° ± 10°C.

C. Reagents (a) Water.—Distilled, deionized, or 18 MΩ resistance for preparation of standard or sample solutions. (b) Standard stock solutions.—Commercially prepared, certified AA standards, or prepared in laboratory by Method 969.23A(c) (see 35.1.21) for Na, 969.23A(d) (see 2.6.01) for K, 965.09B(a) (see 2.6.01) for Ca, 965.09B(b) (see 2.6.01) for Cu, 965.09B(c) (see 2.6.01) for Fe, 965.09B(e) (see 2.6.01) for Mg, 965.09B(f) (see 2.6.01) for Mn, and 965.09B(g) (see 2.6.01) for Zn. (c) Nitric acid.—Unless specified otherwise, use redistilled or ultrapure. (d) Lanthanum oxide.—La2O3, 99.99%; AAS quality. (e) Lanthanum chloride solution.—LaCl3, 1% (w/v). Weigh 11.7 g (± 100 mg) La2O3 and transfer to 1 L volumetric flask. Add enough H2O to wet powder and then slowly add 50 mL concentrated HCl (Caution: exothermic reaction). Let powder dissolve and then dilute to volume with H2O and mix. Lanthanum chloride solution is stable up to 6 months when stored at room temperature. (f) Cesium chloride solution.—CsCl, 10% (w/v). Weigh 12.7 g (±100 mg) CsCl and transfer to 100 mL volumetric flask. Dilute to volume with H2O and mix. Make fresh every 6 months. (g) Filter pulp.—Analyzed ash-free.

D. Ashing Procedure Note: For liquid formulaes, shake container before weighing. Place composite portion in previously cleaned Vycor evaporating dish (which may contain 5 g filter pulp for ease of handling). Exact amount of composite required will depend on concentration of minerals present. (For powders, take ≥1.5 g.) In general, 25 mL will be adequate. If some minerals, in particular Fe, Cu, or Mn, are at very low levels, a larger portion (≤50 mL) may be necessary. Dry portion in 100°C oven overnight or in microwave oven (programmed over ca 30 min). When dry, heat on hot plate until smoking ceases, and then place dish in 525°C furnace (carefully avoiding ignition) for minimum time necessary to obtain ash that is white and free from C, normally 3–5 h, but ≤8 h. Remove dish from furnace and let cool. Ash should be white and free from C. If ash contains C particles (i.e., it is gray), wet with H2O and add 0.5–3 mL HNO3. Dry on hot plate or steam bath and return dish to 525°C furnace 1–2 h. Dissolve ash in 5 mL 1N HNO3, warming on steam bath or hot plate 2–3 min to aid solution. Add solution to 50 mL volumetric flask and repeat with 2 additional portions of 1N HNO3. Dilute to 50 mL with 1N HNO3. (Note: Additional dilutions may be necessary to bring concentrations within the linear range of instrument.)

E. Determination Add LaCl3 solution to final dilution of each standard and test solution to make 0.1% (w/v) La for determination of Ca and Mg only. Add CsCl solution to final dilution of each standard and test solution to make 0.5% (w/v) Cs (0.04M) for determination of Na and K only. Prepare blanks representing all reagents and glassware, and carry through entire procedure. Prepare calibration curve (concentration vs absorbance) for each mineral to be determined, using wavelength and flame specified in Table 985.35B. Optimize flame parameters in accordance with instrument manufacturer’s instructions. Prepare solutions for calibration of instrument to cover linear range of calibration curve. See instrument instruction manual. Assay samples in similar manner. Determine concentration of each mineral from its calibration curve, and calculate concentration in test sample, taking into account test portion size and dilutions. Ref.: J. Assoc. Off. Anal. Chem. 68, 514(1985); J. AOAC Int. 80, 834–844(1997) 986.24 Phosphorus in Infant Formula and Enteral Products, Spectrophotometric Method First Action 1986 Final Action 1988 Revised First Action 1997

Method Performance: See Table 986.24 for method performance data.

A. Principle Phosphorus is determined by spectrophotometry on ashed test portion by complexing with molybdovanadate reagent.

B. Apparatus (a) Spectrophotometer.—Capable of operation at 400 nm. (b) Muffle furnace.—Equipped with pyrometer and controller. (c) Ashing dishes.—Silica or porcelain.

C. Reagents (a) Hydrochloric acid solution.—(1 + 3, v/v). Add 250 mL HCl to 750 mL H2O. (b) Molybdovanadate reagent.—Dissolve 20 g ammonium molybdate in 200 mL hot H2O and cool. Dissolve 1.0 g ammonium metavanadate in 125 mL hot H2O, cool, and add 160 mL HCl. Gradually add, with stirring, molybdate solution to vanadate solution and dilute with H2O to 1.0 L. (c) Phosphorus standard solutions.—(1) Stock standard solution.—2 mg P/mL. Weigh 8.7874 g KH2PO4 previously dried 2 h at 105°C. Quantitatively transfer to 1 L volumetric flask and add ca 750 mL H2O to dissolve. Dilute to volume with H2O. Store in refrigerator. (2) Working standard solution.—0.1 mg P/mL. Dilute 50 mL stock standard solution with H2O to 1 L. Store in refrigerator. Prepare fresh on day of analysis.

D. Preparation of Test Solution Note: For liquid formulae, shake container before weighing. Accurately weigh amount of test portion to contain ca 4.0 mg P into ashing dish and evaporate to dryness on hot plate or steam bath. Ignite in muffle furnace at maximum temperature of 600°C until free of C (3–4 h). Cool; add 40 mL HCl solution, C(a), and several drops of HNO3; and bring to boil on hot plate. Cool, transfer quantitatively to 100 mL volumetric flask, and dilute to volume with H2O.

E. Determination Transfer portions of 0.0, 5.0, 8.0, 10.0, and 15.0 mL working standard solution to respective 100 mL volumetric flasks. These represent 0.0, 0.5, 0.8, 1.0, and 1.5 mg P. Pipet 20.0 mL test solution into each 100 mL volumetric flask. To each standard and test flask, add 20.0 mL molybdovanadate reagent, dilute to volume with H2O, and mix well. Let flasks stand 10 min for complete color development. Determine absorbance of standards and sample in 1 cm cells at maximum near 400 nm. Use 0.0 mg standard to zero spectrophotometer. Use linear regression of standard absorbance vs mg P of standards to determine mg P for each sample.

F. Calculations Caution: See Appendix B: safety notes on safe handling of acids. Dispose of waste solvents in an appropriate manner compatible with applicable environmental rules and regulations.

Calculate content of phosphorus in test sample as follows: P, mg/L = mg P × 5000 × test sample density/g sample

Test sample density should be ca 1.03 g/mL for RTF formula. Ref.: J. Assoc. Off. Anal. Chem. 69, 777(1986); J. AOAC Int. 80, 834–844(1997) Results and Discussion Results for Ca, Mg, Zn, Cu, Fe, Mn, K, Na, and P are summarized in Tables 1–9, respectively. RTF liquid soy formulae and soy powders with similar concentrations of analyte were pooled together as Youden “closely matched pairs” for estimates of within-laboratory variability (RSDr). Despite a different technique in estimating repeatability and reproducibility (RSDR) from that used in the original study, estimates of these parameters can be compared. Use of results from different-day analysis on the same RTF milk-based formulae in the original study produced estimates of repeatability and reproducibility close to those determined in this study. Averaged estimates of repeatability and reproducibility from the original collaborative study are listed with the averaged estimates from this study in Table 10. Exceptions are for Ca and P, which had twice the repeatability and reproducibility estimates as in the original study. Calcium results with the highest variation had the poorest HORRAT ratios (ratio of relative standard deviation among laboratories/relative standard deviation expected, extrapolated from previous studies for analytes at those particular levels). The resulting relative standard deviations estimating betweenlaboratory variation were quite high, and for 2 test samples the HORRAT ratio was more than twice what would be expected at that concentration of analyte. However, no outlier among the Ca results was identified by the Cochran and Grubb’s tests for outliers. The sum ranking test proposed by Youden and Steiner (6) identifies laboratories with consistently high or consistently low results for a significant number of the test samples analyzed. With the sum ranking test applied, 2 laboratories would rank significantly low and one laboratory would rank significantly high. Eliminating the high ranking laboratory results and the low ranking laboratory results for each test sample (see bottom of Table 1) results in relative standard deviations for Ca between laboratories that are closer to the average reproducibility estimate of the original study as listed in Table 10. One collaborative study for Ca, Mg, and P in cheese (7) identified the potential for Ca contamination as a leading factor in between-laboratory variation. In this study, results from 2 laboratories ranked consistently low compared with those reported by other laboratories. No blanks with significant levels of Ca were reported, indicating that Ca contamination was not a problem. One of the low-ranking laboratories weighed a larger amount of test portion and exceeded the 50 g limit recommended by the method for sample 6. The other low-ranking laboratory used different amounts of sample for determining minerals by Method 986.24. This method specifies that liquid formula must be adequately mixed before each weighing to prevent calcium phosphate from settling. The poorest RSDR for Ca was for the 32 fl. oz. can, sample 6. This variation would be

expected for a sample in which 4 times the amount of calcium phosphate could settle to the bottom of the container. The poor RSDR also may be due to greater can-to-can variation. This is indicated by the high RSDR for P for sample 6, which increases to 11% from