Application Note 2007 Hydrolysis Unit E-416, Extraction Unit E-816 Soxhlet Fat determination according to Weibull-Stoldt - Standard application Application No. E-416-E-816-Sox-001, V 1.0 Copyright 2007 Büchi Labortechnik AG 1/8
1 Abstract/Summary A simple and reliable procedure for fat determination of food and feed products according to Weibull-Stoldt is introduced. The sample is hydrolyzed with the Hydrolysis Unit E-416. The Soxhlet extraction is performed with the Extraction Unit E-816. Calculation of total fat content follows gravimetrically after the extract has been dried to a constant weight. This application follows official methods (EN 98/64/EG, AOAC 963.15, 64 / 35 06.00-6). 2 Equipment - Extraction Unit E-816 Soxhlet (SOX) - Hydrolysis Unit E-416 - Mixer B-400 - Analytical balance (accuracy +/- 0.1 mg) - Microwave oven - Drying oven /Vacuum drying oven 3 Chemicals - Quartz sand, particle size 0.3-0.9 mm, Buchi (Order No. 037689) - Celite 545, Macherey-Nagel (815560) - Hydrochloric acid 4 mol/l, 4 L HCl 32% (Hänseler, 20-2000-5) are filled up to 10 L with deionised water - Petroleum ether 40-60 C, analytical grade, Scharlau, EGT (ET 0093) - Diethyl ether, analytical grade, Scharlau, EGT (ET 0081) - Chloroform, puriss.p.a., Fluka (25690) - Hexane, puriss.p.a., Fluka (52765) 4 Samples All samples have been purchased at LGC Promochem (Wesel, Germany) Canned pet (cat) food, certified reference material LGC 7176, specified fat content: 4.46 +/- 0.17 g/100g Madeira cake, certified reference material LGC 7107, specified fat content: 13.4 +/- 0.7 g/100g Processed meat, certified reference material LGC 7152, specified fat content: 15.0 +/- 0.7 g/100g Milk powder (spray dried), certified reference material MUVA RM-55, specified fat content: 26.48 +/- 0.36 g/100 g Chocolate, certified reference material, LGCQM 1003, specified fat content: 29.5 +/- 1 g/100g Milk powder and chocolate (fine powder) did not need any homogenization. The other samples needed homogenization as specified on the certificates: The entire can of pet food was placed directly into the beaker of the Mixer B-400 and mixed once for 2 s. Madeira cake and processed meat were cut into small pieces Application No. E-416-E-816-Sox-001, V 1.0 Copyright 2007 Büchi Labortechnik AG 2/8
(1x1cm), transferred into the beaker of the Mixer B-400 and were mixed 2 to 3 times for 2 s up to a visual homogeneity. 5 Procedure The fat determination according to Weibull-Stoldt includes the following steps: 1. sample homogenization 2. hydrolysis of the sample with 4 M hydrochloric acid to break up the matrix 3. filtration of the hydrolysis solution to separate the fat 4. drying of the filtered sample 5. Soxhlet extraction of the fat 6. drying of the extract 7. weighing of the extract 8. calculation of fat content 5.1 Acid hydrolysis Preparation of the glass sample tubes Add approx. 50 g of quartz sand to the glass sample tube and compact the sand by tapping it gently on the table. Add approx. 5 g Celite 545, spread it evenly by shaking the tubes carefully. The sand and the celite layers should not be mixed together. Otherwise the celite phase may break through the frit and influence the results by a higher recovery or block the frit. Hydrolyzing the sample matrix Place 5 g celite and up to 10 g homogeneous sample 1 in the digestion vessel. Note the accurate weight of the sample. Add 50 ml hydrochloric acid (4 M). Suspend carefully by gently swirling the tube. Add another 50 ml hydrochloric acid (4 M) making sure to rinse any remaining sample off the glass wall. Preheat the Hydrolysis Unit for 10 min. Insert the samples into the unit, lower the vessels, connect the aspiration tubes, reduce the heat to level 3 and start the water jet pump after boiling begins. Violent foaming can be prevented by adding 4 M hydrochloric acid drop by drop. The degree of foaming depends on the sample, but also on the preheating time of the unit. Do not extend preheating excessively. After boiling has been observed, the samples are hydrolyzed for 30 min. At the end of the hydrolysis time, add 100 ml of warm (40-50 C) deionised water to each digestion vessel. Switch off the heating and lift the digestion vessels to the top position in order to filter the hydrolyzate. Wash each of the vessels by gradually adding a total of at least 500 ml warm water, until washed neutral. Check the ph with ph paper. To have the maximum efficiency, aspire all samples/rinsing water at the same time. Stir the celite layers (without touching the sand layer) with a spatula to loosen the pulp. Carefully wipe off the spatula with a piece of tissue and add it on the top of the sample. Dry the glass sample tubes in a vacuum oven ( 4 h at 100 C/200mbar), in an drying oven ( 8 h at 100 C) or in a microwave oven. Using a microwave oven makes the drying much faster. However it is more delicate due to the fact that the 1 The sample weight has to be chosen according to the approximate fat content of the sample. 80-100%: 0.7-1 g 20-50% 1.5-3.5 g <10%: 7-10 g 50-80%: 1-1.5 g 10-20% 3.5-7 g Application No. E-416-E-816-Sox-001, V 1.0 Copyright 2007 Büchi Labortechnik AG 3/8
temperature in the sample can become easily too high (> 105 C) if an inappropriate heating power has been chosen. The following suggestion is valid for the drying of six hydrolyzed samples at the same time. First step: 18 min 640 W, second step: 13 min 480 W (the optimal parameters may depend on the model of microwave). Faster drying at higher temperatures is not recommended because fat can be decomposed at temperatures above 105 C. Oxidized fat can result in an excessive recovery. Allow the glass sample tubes to cool down to room temperature in a desiccator. Add another layer of quartz sand (20 g). This prevents the celite from being resuspended by condensed solvent. 5.2 Fat extraction Preparation of the beakers Always use dry and clean beakers for Soxhlet extraction. Add a boiling aid (eg. boiling stones) to each beaker and dry them for at least 30 min at 105 C. Let them cool down to ambient temperature in a desiccator for at least 1 h. Record the exact weight prior to extraction. Soxhlet extraction Put the sample tubes into the extraction chamber and adjust the level sensor above the level of the sample. See picture 1. Level sensor Picture 1: Soxhlet extraction before start Close the safety shield, lower the rack and fill in the solvent carefully by the condensers. Make sure that the glass sample tubes are not overfilled because Celite, sand or sample particles could overflow into the extractor. The parameters have to be chosen as follows (Table 1). Table 1: Parameters for Soxhlet extraction E-816 Solvent Petroleum ether / Diethyl ether / Chloroform / Hexane 2 Extraction step 120 min (Heater 100%) Rinse step 5 min (Heater 100%) Drying step 15-30 min (Heater 100%) 3 Solvent volume 110-120 ml 4 2 Please select the solvent you are using in the menu. 3 Diethyl ether 15 min, Petroleum ether and Hexane 20 min, Chloroform 30 min. 4 Solvent volume depends on the sample (voluminous samples absorb more solvent) and on the setting of the level sensor. Generally 110 ml are suitable for most applications. Application No. E-416-E-816-Sox-001, V 1.0 Copyright 2007 Büchi Labortechnik AG 4/8
Drying of the extract Dry the beakers containing the extract in a drying oven at 103 C to constant weight. Let the beakers cool down to ambient temperature for at least 1 h in a desiccator and record the weight. Make sure, that the cooling down time of the beakers in the dessicator is the same before and after extraction. Differences in beaker's temperature falsify the results. 6 Calculation The results are calculated as percentage of fat using equation (1). ( mtotal mbea ker ) % Fat = 100% (1) m sample %Fat : percentage of fat in the sample m total : beaker weight + extract [g] m beaker : empty beaker weight [g] m sample : sample weight [g] 7 Results Table 2: Canned Pet food (Specification 4.46 +/- 0.17 g/100g) S 1 4.45 4.58 4.61 4.59 S 2 4.47 4.56 4.63 4.58 S 3 4.49 4.30 5 4.61 4.62 S 4 4.50 4.56 4.68 4.60 S 5 4.47 4.56 4.64 4.59 S 6 4.47 4.56 4.68 4.61 Mean value g/100g 4.48 4.57 4.64 6 4.60 rsd [%] 0.39 0.19 0.64 0.20 Table 3: Madeira cake (Specification 13.4 +/- 0.7 g/100 g) S 1 12.93 13.20 13.43 13.23 S 2 12.83 13.12 13.59 13.24 S 3 12.95 13.19 13.43 13.20 S 4 12.91 13.21 13.49 13.26 S 5 13.11 13.23 13.38 13.20 S 6 13.01 13.23 13.44 13.41 Mean value g/100g 12.96 13.20 13.46 13.26 rsd [%] 0.74 0.34 0.54 0.59 5 this value is an outlier according to the statistical test of Dean-Dixon for outliers (α=5%) 6 result not within the specifications, Chloroform has often larger amount extract than the other solvents Application No. E-416-E-816-Sox-001, V 1.0 Copyright 2007 Büchi Labortechnik AG 5/8
Table 4: Processed meat (Specifications: 15.0 +/- 0.7 g/100g) S 1 15.35 15.30 15.31 15.24 S 2 15.02 15.32 15.54 15.33 S 3 15.07 15.47 15.47 15.35 S 4 15.26 15.40 15.54 15.25 S 5 15.25 15.32 15.44 15.18 S 6 15.23 15.30 15.43 15.26 Mean value g/100g 15.20 15.35 15.45 15.27 rsd [%] 0.84 0.45 0.55 0.41 Table 5: Milkpowder (Specifications: 26.48 +/- 0.36 g/100g) S 1 26.45 26.57 26.61 26.18 S 2 26.47 26.53 26.63 26.36 S 3 26.53 26.50 26.53 26.38 S 4 26.58 26.56 26.81 26.42 S 5 26.57 26.47 27.21 26.31 S 6 26.45 26.20 26.62 26.38 Mean value g/100g 26.51 26.47 26.74 26.34 rsd [%] 0.22 0.53 0.93 0.32 Table 6: Chocolate (Specifications: 29.5 +/- 1.0 g/100g) S 1 29.67 29.92 30.30 29.90 S 2 29.65 29.92 30.19 29.77 S 3 29.31 29.85 30.30 29.87 S 4 29.24 29.85 30.30 29.74 S 5 29.55 29.76 30.44 29.78 S 6 29.52 29.92 30.59 29.80 Mean value g/100g 29.49 29.87 30.35 29.81 rsd [%] 0.61 0.21 0.46 0.21 Depending on the type of solvent used, the amount of extract differs because of the different polarity of the solvents which affects the mass transfer to a great extend, resulting in different results. The determined fat content of the certified reference materials corresponded to the specified values. The relative standard deviations (rsd) are very low, being < 1% for all samples. Application No. E-416-E-816-Sox-001, V 1.0 Copyright 2007 Büchi Labortechnik AG 6/8
8 Comparison to official methods HCl Solution in Hydrolysis Hydrolysis time Extraction time Büchi's Application note EN 98/64/EG AOAC 963.15 64 / 35 06.00-6 Explanation 4 M HCl 3M HCl 4.4 M HCl 4 M HCl no influence (100 ml) (100 ml) (100 ml) (150 ml) 30 min 60 min 15 min 60 min 30 min seems to be an appropriate time for most of the samples 2 h 6 h 4 h (30 cycles) 4 h Automated Soxhlet as E-816 has the advantage of a faster extraction because cycle rate is higher than with the classic glass ware. 9 Troubleshooting High recovery - Drying of the extract was not sufficient. Dry to a constant weight - Decomposition of the fat due to too high drying temperatures during drying step. Some fats and oils are very heat sensitive (eg. Sunflower oil). Lower the drying temperature by decreasing the percentage of the heater and dry the extracts at low temperatures and under reduced pressure in a vacuum oven. - Celite has been washed out: Please loosen the pulp carefully before drying the sample. - Use clean solvent. When emptying the tank, make sure, that the outer part of the flexible tube (Tygon) does not come into contact with the solvent, plasticizer could be dissolved out. Low recovery Loss of the sample during hydrolysis step: - Please wash the digestion vessels carefully, so that the entire sample is transferred into the sample tubes. - Use the rinsing water at the recommended temperature: Too warm water can cause a loss of the fat, too cold an insufficient rinsing of the digestion vessel and aspiration tube. Incomplete extraction: - Level sensor was set too high (less cycles) or too low (sample was not soaked completely in the solvent - Too short extraction time was chosen: Please use the recommended parameters. (Table 1) - Accumulation of solvent on the top of the sample, leaving less solvent for the extraction cycle can cause insufficient extraction, see paragraph "Accumulation of solvent". Different beaker temperature: - The warmer the beakers, the less weight they have. If you weigh the beakers before and after extraction after different cooling down times, you get a too low/too high recoveries. Let the beakers cool down to ambient temperature and weigh them again. Too large variation - Too small sample volume: Please use the recommended sample weights, if a sample is very inhomogeneous (e.g. Salami), increase sample weight. - Insufficient homogenization of the sample. Please use an appropriate mixer (eg. Mixer B-400) and/or a mortar and pistil for homogenization - Incomplete extraction can also be the reason for a large variation. Please consider the paragraph "low recovery". Application No. E-416-E-816-Sox-001, V 1.0 Copyright 2007 Büchi Labortechnik AG 7/8
- Low fat content of the sample lead generally to higher rsd. Boiling retardation - Always use boiling aids / stones. - When celite, dust or particles from the sample has been washed out, boiling retardation can occur. Make sure the celite and sand layers are not mixed up and celite and sand have the correct particle size. With some samples and solvents (chloroform) more particles/celite is washed out than with others. To prevent boiling retardation, decrease heater % and increase time of the drying step. Clean the extraction chamber on a regular basis. Accumulation of solvent on top of the sample - The sample has been dried insufficiently. - The heating level has been set too high: Verify that the heater is set to 100% and that the solvent has been chosen correctly. Depending on the sea level and the ambient temperature, you may have to adjust heater percentage. - The celite layers have not been loosened prior to extraction. - The frit of the sample tube is blocked: Please clean the sample tube and/or replace it. 10 References Operation Manuals - Mixer B-400 - Hydrolysis Unit B-411/E-416 - Extraction Unit E-816 Sox Official methods - Commission directive 98/64/EC analysis for the determination of amino-acids, crude oils and fats, and olaquindox in feedingstuffs. - AOAC Official methods 963.15 Fat in Cacao Products - 64 (formerly 35) 06.00-6 Bestimmung von Gesamtfett in Fleisch und Fleischerzeugnissen BÜCHI Labortechnik AG CH-9230 Flawil 1/Switzerland T +41 71 394 63 63 F +41 71 394 65 65 www.buchi.com Quality in your hands Application No. E-416-E-816-Sox-001, V 1.0 Copyright 2007 Büchi Labortechnik AG 8/8