
Article 1 
Annex I to Regulation (EC) No 900/2008 is replaced by the text set out in the Annex to this Regulation.
Article 2 
This Regulation shall enter into force on the 20th day following its publication in the Official Journal of the European Union.
This Regulation shall be binding in its entirety and directly applicable in all Member States.Done at Brussels, 9 February 2010.
For the Commission
The President
José Manuel BARROSO
ANNEX


ANNEX I  1. 
This method describes the determination of the content of starch and its degradation products including glucose in food products for human consumption hereafter referred to as “starch”. The starch content is determined from the quantitative analysis of glucose by high-performance liquid chromatography (HPLC) after enzymatic conversion of starch and its degradation products into glucose.
 2. 
The total glucose content means the value Z as calculated in point 7.2.1 of this Annex. It represents the content of starch and all its degradation products, glucose included.

The starch/glucose content as defined in Annex III to Regulation (EC) No 1460/96 shall be calculated on the basis of the total glucose content Z and as set out in Article 2 point 1 of this Regulation.

The starch (or dextrin) content as referred to in column 3 of Annex IV to Commission Regulation (EC) No 1043/2005 shall be calculated on the basis of the total glucose content Z as set out in Article 2 point 2.1 of Commission Regulation (EC) No 904/2008.

The Starch content referred to in point 1 of this Annex means the value E, as calculated in point 7.2.2 of this Annex. It is expressed in % (m/m). It is equivalent to the total glucose content Z, expressed as starch. This value E does not interfere in the above mentioned calculations.
 3. 
The samples are homogenised and suspended in water. The starch and its degradation products, present in the samples, are enzymatically converted into glucose in two steps:


1.. Starch and its degradation products are partially converted into soluble glucose chains using thermostable alpha-amylase at 90 °C. For effective conversion it is necessary that the samples should be completely solved or should be present in the form of a suspension containing very small solid parts
2.. The soluble glucose chains are converted into glucose using amyloglucosidase at 60 °C.

Products containing a high content of proteins or fat are clarified and filtrated.

The determination of sugars is performed by HPLC analysis.

Because a partial inversion of sucrose may occur during the enzymatic treatment, the determination of free sugars is also performed by HPLC analysis to calculate the corrected glucose content.
 4. 
Use reagents of recognised analytical grade and demineralised water.
 4.1. Glucose, min 99 %.
 4.2. Fructose, min 99 %.
 4.3. Sucrose, min 99 %.
 4.4. Maltose-monohydrate, min 99 %.
 4.5. Lactose-monohydrate, min 99 %.
 4.6. Solution of thermostable alpha-amylase (1,4-alpha-D-Glucan-glucanohydrolase), with activity about 31 000 U/ml (1U will liberate 1,0 mg of maltose from starch in 3 minutes at pH 6,9 and 20 °C). This enzyme can contain a low amount of impurities (e.g. glucose or sucrose) and other interfering enzymes. Storage at ca. 4 °C. Alternatively, other sources of alpha-amylase may be used yielding a final solution with comparable enzyme activity.
 4.7. Amyloglucosidase (1,4-alpha-D-Glucan glucohydrolase) from Aspergillus niger, powder with activity about 120 U/mg or about 70 U/mg (1U will liberate 1 micromol glucose from starch per minute at pH 4,8 and 60 °C). This enzyme can contain a low amount of impurities (e.g. glucose or sucrose) and other interfering enzymes (e.g., Invertase). Storage at ca. 4 °C. Alternatively, other sources of amyloglucosidase may be used yielding a final solution with comparable enzyme activity.
 4.8. Zinc acetate dihydrate, p.a..
 4.9. Potassium hexacyanoferrate (II) (K4[Fe(CN)]6.3H2O), extra pure.
 4.10. Sodium acetate anhydrous, p.a..
 4.11. Glacial acetic acid, 96 % (v/v) (minimum).
 4.12. Sodium acetate buffer (0,2 mol/l). Weigh 16,4 gram sodium acetate (point 4.10) into a beaker glass. Dissolve in water and rinse into a volumetric flask of 1 000 ml. Dilute to the mark with water and adjust the pH to 4,7 with acetic acid (by use of a pH-meter (point 5.7). This solution may be used for max 6 months with storage at 4 °C.
 4.13. Amyloglucosidase solution. Prepare a solution of amyloglucosidase powder (point 4.7) by using sodium acetate buffer (point 4.12). The enzyme activity must be sufficient and in accordance with the starch content in the amount of sample (for example, activity about 600 U/ml is obtained from 0,5 g amyloglucosidase powder 120 U/mg (point 4.7) in a final volume of 100 ml for 1 g starch in the amount of sample). Prepare immediately before use.
 4.14. Reference solutions. Prepare solutions of glucose, fructose, sucrose, maltose and lactose in water, as conventionally used in the HPLC analysis of sugars.
 4.15. Reagent for clarification (Carrez I). Dissolve 219,5 gram zinc acetate (point 4.8) in water in a beaker glass. Rinse into a volumetric flask of 1 000 ml and add 30 ml acetic acid (point 4.11). Mix thoroughly and dilute to the mark with water. This solution may be used for max 6 months while stored at ambient temperature. Other clarification reagents, equivalent to Carrez solution, may be used.
 4.16. Reagent for clarification (Carrez II). Dissolve 106,0 gram potassium hexacyanoferrate (II) (point 4.9) in water in a beaker glass. Rinse into a volumetric flask of 1 000 ml. Mix thoroughly and dilute to the mark with water. This solution may be used for max. 6 months while stored at ambient temperature. Other clarification reagents, equivalent to Carrez solution, may be used.
 4.17. HPLC Mobile phase. Prepare a mobile phase which is conventionally used in the HPLC analysis of sugars. In case of using an aminopropyl silicagel column, e.g., a common mobile phase is a mixture of HPLC grade water and acetonitrile.
 5.  5.1. Standard laboratory glass ware.
 5.2. Fluted filters, e.g., 185 mm.
 5.3. Syringe filters, 0,45 μm, suitable for aqueous solutions.
 5.4. Sample vials suitable for the HPLC autosampler.
 5.5. 100 ml volumetric flasks.
 5.6. Plastic syringes, 10 ml.
 5.7. pH-meter.
 5.8. Analytical balance.
 5.9. Water bath with thermostat, adjustable to 60 °C and 90 °C.
 5.10. HPLC Apparatus suitable for analysis of sugars.
 6.  6.1. 
The product is homogenised.
 6.2. 
The amount of sample is estimated from the ingredient declaration and the conditions of the HPLC analysis (concentration of the glucose reference solution), and shall not exceed:

amount of sample (g) =volume of volumetric flask (e.g., 100 ml)estimated starch content (%)

Weigh the sample to 0,1 mg accuracy.
 6.3. 
The blank is determined by performing a complete analysis (as described in point 6.4), without adding sample. The result of the blank determination is used in the calculation of the starch content (point 7.2).
 6.4.  6.4.1. 
Homogenise the sample by shaking or stirring. The chosen test portion (point 6.2) is weighed into a volumetric flask (point 5.5) and about 70 ml warm water is added.

After dissolving or suspending, add 50 microliter of thermostable alpha-amylase (point 4.6) and heat at 90 °C for 30 min in a water bath (point 5.9). Cool as quick as possible to 60 °C in a water bath, and add 5 ml of a solution of amyloglucosidase (point 4.13). For samples which could influence the pH of the reaction solution, control the pH and adjust it to 4,6 to 4,8, if necessary. Allow to react for 60 min at 60 °C. Cool the samples to ambient temperature.
 6.4.2. 
For samples with a high content of proteins or fat, clarification is necessary by adding 1 ml Carrez I (point 4.15) to the sample solution. After shaking, 1 ml Carrez II (point 4.16) is added. Shake the sample again.
 6.4.3. 
The sample in the volumetric flask is diluted to the mark with water, homogenised and filtered through a fluted filter (point 5.2). Collect the sample extract.

Filter the extracts through a syringe filter (point 5.3) with a syringe (point 5.6) that has been preflushed with the extract. Collect the filtrates in vials (point 5.4).
 6.5. 
HPLC is performed as conventionally for analysis of sugars. If the HPLC analysis shows traces of maltose, then the starch is incompletely converted, which results in a too low recovery for glucose.
 7.  7.1. 
For the calculation of the starch content, the results of two HPLC analysis are necessary, namely sugars present in the sample before (‘free sugars’) and after enzymatic treatment (as described in this method). Also a blank determination has to be performed to be able to correct for sugars present in the enzymes.

In the HPLC analysis, the peak area is determined after integration and the concentration is calculated after calibration with reference solutions (point 4.14). From the glucose concentration (g/100 ml) after enzymatic treatment, the concentration of glucose (g/100 ml) in the blank is subtracted. Eventually the content (g sugar/100 g sample) of sugars is calculated using the weighted amount of sample, which results in:


1.. HPLC analysis before enzymatic treatment, giving the content (g/100 g) of free sugars:

— glucose G
— fructose F
— sucrose S
2.. HPLC analysis after enzymatic treatment, giving the content (g/100 g) of sugars:

— glucose after correction for the blank (Ge cor)
— fructose Fe
— sucrose Se
 7.2.  7.2.1. 
If the amount of fructose after enzymatic treatment (Fe) is higher than the amount of fructose before enzymatic treatment (F), then the sucrose, present in the sample, is partly converted into fructose and glucose. This means that a correction shall be made for the liberated glucose (Fe – F).

Z, final glucose content after correction in g/100g:

Z = (Ge cor) – (Fe – F)
 7.2.2. 
E, “starch” content in g/100g:

E = [(Ge cor) – (Fe – F)] × 0,9
 8. 
Details of an inter laboratory test relating to precision data of the method performed on 2 samples are given in this point. They reflect the performance requirements for the method described in this annex.

An inter laboratory test was carried out in 2008 with the participation of the European Customs laboratories.

The evaluation of precision data was performed according to the “Protocol for the design, conduct and interpretation of method-performance studies”, W. Horwitz, (IUPAC technical report), Pure & Appl. Chem., Vol. 67, No 2, PP.331-343, 1995.

The precision data are given in the table below.


Samples1chocolate and biscuit bar2biscuit Zsample 1 Zsample 2
Number of laboratories 41 42
Number of laboratories after eliminating outliers 38 39
Mean (%, m/m) 29,8 55,0
Repeatability standard deviation sr (%, m/m) 0,5 0,5
Reproducibility standard deviation sR (%, m/m) 1,5 2,3
Repeatability limit r (%, m/m) 1,4 1,4
Reproducibility limit R (%, m/m) 4,2 6,6

