The Fluorometric Determination of Vitamin A in Dairy Products

Health Protection Branch
Bureau of Nutritional Sciences
Ottawa

Laboratory Procedure LPFC-200
December 1986

Help on accessing alternative formats, such as Portable Document Format (PDF), Microsoft Word and PowerPoint (PPT) files, can be obtained in the alternate format help section.

---

Definition:

This method is applicable to the determination of vitamin A in skim milk, evaporated skim milk, evaporated partly skimmed milk, skim milk powder, flavoured skim milk, flavoured partly skimmed milk, condensed skim milk and other skim milk or partly skimmed milk products in accordance with Sections B.08.004, 005, 011, 012, 014, 017, 018, 019, 020, 023 and 026 of the Food and Drug Regulations.

Scope:

This method has been used extensively to measure the vitamin A content of skim milk, partly skimmed milk and skim milk and partly skimmed milk related products. Precision, accuracy and recovery data are shown in Reference 1 and in Tables 1 to 4.

THE ANALYST SHOULD BECOME FAMILIAR WITH THIS METHOD USING STANDARDS AND/OR SPIKED SAMPLES BEFORE UNDERTAKING ANALYSES OF UNKNOWNS.

Principle:

Samples are saponified to convert retinol esters to retinol and to destroy interfering substances. Free retinol is extracted with hexane and is measured by fluorescence spectrophotometry (1) (2).

Caution:

Chemicals represent a variety of potential hazards and must be handled with care. For details respecting chemical hazards, refer to the precautionary notes described in Chapter 51, Laboratory Safety, Official Methods of Analysis, Association of Official Analytical Chemists, Arlington, VA, Fourteenth Edition, 1984 and other recognized texts respecting laboratory safety.

Apparatus:

1. Fluorescence spectrophotometer, Perkin Elmer LS-5 or equivalent;
   
   
2. Bench centrifuge;
   
   
3. Controlled temperature water bath at 80oC;
   
   
4. Automatic pipette, Oxford Micro-pipetting System or equivalent, set at 1, 2 and 5mL;
   
   
5. Saponification tubes:
   
   
   1. 15 mL graduated centrifuge tubes with ground glass stoppers,
      
      
   2. a ground glass joint (inner part 14/35 Pyrex No. 6560), as purchased, with a stem 10 cm, long, used as an air condenser;
      
      
6. Vortex mixer;
   
   
7. Nitrile latex gloves, Fisher No. 11-394-23;
   
   
8. Pasteur pipettes, disposable, with rubber bulb; 9. Security goggles;
   
   
9. Volumetric pipetter, 1.0 and 2.0 mL for dilution of standard

Reagents:

1. U.S.P. Vitamin A Reference Standard, United States Pharmacopeial Convention, Inc., Rockville Md. 20852;
   
   
2. Hexane, HPLC grade;
   
   
3. Ethanol, 95%, or anhydrous,
   
   
   1. distill from solid AgNO₃ (ca 2.5 g/100 mL) and KOH pellets (ca 5.0 g/100 mL); discard first portion and collect when distilling at 78.5^oC;
      
      
   2. store in a glass-stoppered bottle;
      
      
4. Pyrogallol, 1% w/v solution in ethanol (Reagent 3);
   
   
5. KOH pellets (ca 100 mg pellets)
   
   
6. Antifoaming agent, Antifoam. FG-10 or equivalent.

Procedure:

Laboratory should be illuminated with gold fluorescent or tungsten lamps to
avoid destruction of vitamin A.

A. Analysis:

I. Preliminary Preparations;

1. Distill ethanol from KOH + AgNO3 (Reagent 3);
   
   
2. Prepare ethanolic pyrogallol (l% w/v) - must be freshly prepared;
   
   
3. Heat water bath to 80oC;
   
   
4. Rinse with hexane one air condenser and one centrifuge tube for
   each sample and blank;
   
   
5. Switch on fluorescence spectrophotometer 30 min before taking
   readings;
   
   
6. Set fluorescence spectrophotometer at 330 nm excitation and 480 nm emission with both slits at l0nm band pass (8 nm band pass for other than the LS-5);
   
   
7. Check that cells are clean and dry; if necessary, wash in soapy water and rinse with: distilled water, then alcohol and finally hexane.
   
   II. Preparation of Standards:
   
   
8. Cut tip of U.S.P. Vitamin A Reference Standard and squeeze oil solution into a weighed 25 mL graduated flask; reweigh to determine weight (± 1 mg) of oil solution (usually ca 220 mg);
   
   
9. Dissolve the oil solution in hexane, dilute to volume, mix well and label "Solution A";
   
   
10. Dilute 1.0 mL of Solution A to 10 mL with hexane using a 10 mL volumetric flask, mix well and label "Solution B";
    
    
11. Dilute 1.0 mL of Solution B to 50 mL with 1% w/v ethanolic pyrogallol (Reagent 4) using a 50 mL volumetric flask, mix well and label "Solution C";
    
    
12. Pipette 1.0 mL distilled water, 2.0 mL Solution C and add 0.5 g KOH (5 pellets, Reagent 5) into 3 saponification tubes;
    
    
13. Saponify and extract as directed for the samples beginning Section IV, step 34;
    
    
14. To prepare a standard curve, dilute 1-5 mL Solution C to 10 mL with Reagent 4 and proceed as directed for the standard (Section IV, step 34).
    
    III. Preparation of Samples:
    
    (a) Skim or Partly Skimmed Fluid Milk:
    
    
15. Allow samples to reach room temperature (22oC);
    
    
16. If sample(s) consists of only one container, shake before opening then pipette aliquots for analysis;
    
    
17. If sample(s) consists of more than one container then pipette aliquots from each container or pool 100 mL aliquots in an appropriate beaker;
    
    
18. Stir thoroughly with a glass rod but avoid the production of foam;
    
    
19. Pipette 1.0 mL portions for analysis immediately after mixing.
    
    (b) Condensed or Evaporated Skim or Partly Skimmed Milk:
    
    
20. Prepare as directed for "Skim or Partly Skimmed Fluid Milk" but dilute 2 volumes to 5 volumes before analysis;
    
    
21. If the label declaration is in IU per 100 g, measure the density with a densitometer or dilute 20 g to 50 mL for analysis;
    
    (c) Skim or Partly Skimmed Milk Powders:
    
    
22. Mix sample in a plastic bag;
    
    
23. Weigh 100 g and transfer to a 1000 mL graduated flask;
    
    
24. Add 1 g antifoaming agent (e.g. Antifoam FG-10);
    
    
25. Add Teflon coated magnetic stirring bar (3 cm x 1 cm diam, volume ca 3 mL);
    
    
26. Add distilled water, flush flask with nitrogen and stir until homogeneous;
    
    
27. Dilute to the mark with distilled water and allow to stand until volume does not change;
    
    
28. Dilute to the mark again;
    
    
29. Repeat steps 27-28 until no change;
    
    
30. Mix well by inverting flask and immediately take aliquots for analysis;
    
    
31. Depending on the label declaration, it may be necessary to measure the density of the powder and the reconstituted milk;
    
    
32. For purposes of this method, the average density of the reconstituted milk can be assumed to be 1.03;
    
    IV. Saponification and Extraction:
    
    
33. Using automatic pipettes for liquids, add the following to each saponification tube:
    
    
    1. 0.5 g (5 pellets) KOH,
    2. 2.0 mL ethanolic pyrogallol, 1% w/v,
    3. 1.0 mL sample or 1.0 mL distilled water for blank;
       
       
34. Mix thoroughly with Vortex mixer (ca 30 sec);
    
    
35. Attach air condenser;
    
    
36. Place in water bath at 80oC for 20 min;
    
    
37. The contents should reflux gently. (There should be no substantial reduction in volume. If this occurs, the tube has been over heated and should be rejected.);
    
    
38. Add 2 mL of distilled water;
    
    
39. Remove the condensers;
    
    
40. Stopper the tubes;
    
    
41. Allow to cool in the dark (The samples may be stored at room temperature overnight under nitrogen at this stage.);
    
    
42. Add 5 mL of hexane;
    
    
43. Return stoppers;
    
    
44. Mix thoroughly for 30 seconds using a Vortex mixer;
    
    
45. Centrifuge at ca 1000 rpm for 3 min;
    
    
46. Transfer upper hexane layer, using a Pasteur pipette, to a 10 mL volumetric flask;
    
    
47. Repeat the extraction procedure with an additional 5 mL of hexane;
    
    
48. Combine the extracts;
    
    
49. Make up to 10 mL with hexane and mix well;
    
    
50. Take readings with fluorescence spectrophotometer within 1 hour;
    
    V. Measurement of Fluorescence Using a Fluorescence Spectrophotometer:
    
    
51. Rinse cells with hexane;
    
    
52. Allow to drain;
    
    
53. Read samples, standards and blanks (which should be 2-5 units) and hexane (which should be ca 1 unit);
    
    VI. Calculation of Results:
    
    
54. Subtract the value of the blank from the values for the samples and standards;
    
    
55. Calculate equivalent values of standards as follows;
    
    Equivalent value (Units/100 mL sample) - (W x P)/62.5 where
    W = weight of oil solution in mg,
    P = potency of oil solution (Units/mg) as declared on label of the U.S.P.Vitamin A Reference Standard;
    
    
56. Calculate values for samples by comparison of readings with those of standards.

References:

1. The Fluorometric Determination of Vitamin A in Dairy Products. Thompson, J.N., Erdody, P., Maxwell, W.B. and Murray, T.K., J. Dairy Sci., 55, 1077, (1972).
   
   
2. Destruction by Light of Vitamin A in Dairy Products, Thompson, J.N. and Erdody, P. J. Can. Inst. Food Sci. and Technol. 7, 157, (1974).
   
   
3. Chapter on Vitamin A Parrish, D.B., Moffitt, R.A., Noel, R.J. and Thompson, J.N. Methods of Vitamin Assay (p. 153-184) Augustin, J., Klein, B.P., Becker, D.A. and Venugopal, P.B.eds. Fourth Edition, 1985 John Wiley and Sons, New York

Method Evaluation

Accuracy

All milk contains vitamin A from natural sources, which is measured with the vitamin added by dairies. It is difficult, therefore, to produce a sample of known composition to test the validity of methods. The accuracy of Method LPFC-200 must be estimated indirectly from investigation of potential sources of error. Only traces of interfering substances have been observed during chromatography of extracts of milk prepared as in the method. Small amounts of fluorescent lipid are always produced from contaminants in the digesting ethanol, and they are measured as the "blank". The "blank" values are usually significantly reduced when ethanol is purified before use as described in the method. Large amounts of fluorescent materials are produced, however, when the digest is overheated. Comparison of the results of Method LPFC- 200 with those of a recent (1986) HPLC method suggest that the fluorometric method overestimates the true retinol content by about 10% (Table 1).

Table 1 - Comparison of Results of Method LPFC-200 with HPLC Measurement of Retinol in Eleven Samples of Fortified Milk

View Comparison of Results of Method LPFC-200 with HPLC Measurement of Retinol in Eleven Samples of Fortified Milk Table

Recovery

Recovery should not be measured by adding pure vitamin A to milk. The
vitamin does not disperse easily, it is rapidly oxidized and it sticks to
glassware. The vitamin can be added in the form of commercial dispersible preparations, but these materials must be examined by another method to determine how much vitamin has been added. A test of this nature is described in reference (1).

Precision

The variations observed within a laboratory in the results of repeated analysis of the same sample are shown in Table 2.

Table 2 - Variation in One Laboratory in Replicate Determinations

View Variation in One Laboratory in Replicate Determinations Table

The vitamin should be evenly distributed in milk and variation is not usually
increased when different samples of a production lot are examined (Table 3).

Table 3 - Variation in Determinations in One Laboratory on Different Samples from Single Production Lots

View Variation in Determinations in One Laboratory on Different Samples from Single Production Lots Table

Variation among laboratories in routine analysis is shown in Table 4.

Table 4 - Variation Among Laboratories in Milk Check Samples
(Units/100 mL)

View Variation Among Laboratories in Milk Check Samples (Units/100 mL) Table