Food Science & Technology · Food Science Core · also Food Analysis, Analytical Techniques in Food, Food Chemistry Laboratory, Instrumental Methods of Analysis
How a food laboratory measures what is in a food: sampling, proximate analysis, official methods, and the instruments — spectroscopy, chromatography, mass spectrometry and electrophoresis — that a food analyst runs. It is the core of the food analyst qualification in countries that license analysts.
In the exams: ICAR AIEEA-PG Code 18 Unit I (modern analytical techniques in food analysis); ICAR AICE 35.b Unit 6 (research techniques); core of food analyst examinations
Food analysis answers the questions every part of the food system asks: what is in this food, how much, and does it meet its specification, its label and the law? Producers need composition to formulate and label products and to control processes; buyers need it to check that they receive what they pay for; regulators need it to detect adulteration, contaminants and misbranding; nutritionists and researchers need it to build food composition tables and study diet and health. A food analyst combines chemistry, instrumentation and statistics with an understanding of how foods are made, because the right method depends on the matrix — a method that works for flour may fail for chocolate or cheese.
This subject covers the core of food analysis: sampling and sample preparation; official methods and the compendia that publish them; proximate analysis — moisture, ash, protein, fat and carbohydrate, including dietary fibre; mineral and vitamin analysis; the main instrumental techniques — UV-visible, fluorescence and infrared spectroscopy, atomic absorption and inductively coupled plasma spectroscopy, thin-layer, gas and liquid chromatography, mass spectrometry, electrophoresis and immunoassays; common physical tests such as pH, titratable acidity, refractometry and colour; and the validation and quality control that make results trustworthy. The chemistry of the constituents being measured is explained in food-chemistry-major-constituents and food-chemistry-minor-constituents-enzymes; instruments for physical properties, rheology and colour are covered in more depth in food-physical-properties-rheology; microbiological methods are in food-microbiology; and the detection of adulterants and contaminants builds on this subject in food-fraud-authenticity and food-toxicology-contaminants-allergens.
Food analysis is examined in ICAR AIEEA-PG Code 18 (modern analytical techniques in food analysis), ICAR AICE-JRF/SRF (research techniques: spectroscopy, chromatography, electrophoresis, ultracentrifugation and potentiometry), and above all in the qualifying and recruitment examinations for food analysts and food safety officers, such as India's Food Analyst examination conducted by the food regulator, where method principles, calculations and laboratory practice are central. It is also the daily work of quality-control and regulatory laboratories accredited to ISO/IEC 17025. Method details summarised here must be taken, in practice, from the current official method text.
An analysis can be no better than the sample it is made on. The aim of sampling is to obtain a portion that represents the lot — a consignment, a production batch, a tank — for the property being measured. Foods are often heterogeneous: fat separates in milk, fines segregate in powders, contaminants such as aflatoxins occur in a few highly contaminated kernels, and composition varies between the surface and the centre of a carcass or cheese. Sampling plans therefore specify how many units to take, from where, and how to combine them. Random sampling gives every unit an equal chance of selection; stratified sampling divides the lot into layers or groups and samples each; systematic sampling takes units at regular intervals. For legal enforcement, the procedure — number of units, sealing, labelling, division into parts for the laboratory, the food business and possible appeal, and the time limits for dispatch — is laid down in food law and must be followed exactly, or results may not stand in court.
The bulk or aggregate sample collected in the field is reduced to a laboratory sample and then to the test portion actually analysed. Reduction must keep representativeness: coning and quartering (heaping the material, flattening it, dividing it into quarters and keeping opposite quarters) and riffle dividers are used for granular materials. Sampling error is often larger than analytical error, particularly for contaminants distributed unevenly, which is why regulatory sampling plans for mycotoxins specify large aggregate samples and grinding of the whole sample before sub-sampling.
Sample preparation turns the laboratory sample into a homogeneous test portion suited to the method, without changing the analyte. Steps include grinding, blending or homogenising; for moist foods, care is needed to avoid moisture loss or gain; for fatty foods, to avoid fat separation; for oxidisable or light-sensitive analytes such as vitamins, to work quickly, in subdued light, sometimes under nitrogen and with antioxidants. Enzyme activity may need to be stopped by freezing or heating. Samples are stored at temperatures that preserve the analyte — often frozen — in inert containers, and documented through a chain of custody.
Extraction and clean-up separate the analyte from the matrix: solvent extraction, solid-phase extraction, liquid–liquid partitioning, and for pesticide residues the widely used QuEChERS approach (quick, easy, cheap, effective, rugged and safe), combining acetonitrile extraction with salting-out and dispersive solid-phase clean-up. Digestion with acids or by dry ashing releases minerals; hydrolysis with acid, alkali or enzymes releases bound fat, vitamins or fibre components. Each step is a potential source of loss or contamination, which is why recoveries are checked during validation.
Official methods are methods that have been collaboratively studied — tested in many laboratories on the same samples — and adopted by an authoritative body, so that results are comparable and defensible. The most widely used compendium is the Official Methods of Analysis of AOAC INTERNATIONAL, whose methods carry numbers indicating the year of adoption (for example, AOAC 985.29 for total dietary fibre). The International Organization for Standardization (ISO), often jointly with the International Dairy Federation (IDF) for dairy methods, publishes international standard methods; national bodies such as standards institutes and food regulators publish their own manuals — for example the FSSAI Manuals of Methods of Analysis of Foods in India and the methods of the American Association of Cereal Chemists (now Cereals & Grains Association) and the American Oil Chemists' Society for cereals and oils.
The Codex Alimentarius classifies methods for checking compliance with its standards into four types. Type I (defining) methods are those that define a value that can only be obtained by that method, such as some empirical methods for moisture or crude fibre. Type II (reference) methods are designated where Type I does not apply and are used in disputes. Type III (alternative approved) methods meet the performance criteria and may be used for control, inspection or regulation. Type IV (tentative) methods are traditional or recent methods not yet fully validated. Increasingly, Codex and regulators set performance criteria — trueness, precision, limits of detection and quantification — rather than prescribing a single method, so that laboratories can use any method that is shown to meet them.
Moisture content is the most frequently measured property of foods because it affects stability, texture, yield and price, and because other results are often expressed on a dry basis. The commonest methods measure weight loss on drying. In air-oven methods, a weighed test portion is dried at a specified temperature — often about 100 to 105 °C — for a set time or to constant weight, and moisture is calculated as the loss in mass: moisture (%) = (initial mass − dry mass)/initial mass × 100. For example, a 5.000 g sample that weighs 4.350 g after drying has lost 0.650 g, or 13.0 per cent moisture. Vacuum ovens dry at lower temperatures, commonly around 70 °C under reduced pressure, which suits sugar-rich products such as honey, dried fruit and jams that would decompose or caramelise at higher temperatures.
Drying methods are empirical: they measure all volatile matter lost under the stated conditions, not water alone, and they may not remove tightly bound water. Volatile compounds (in spices, for example) are counted as moisture; fat oxidation can gain weight; sugars can decompose and lose water of their own. The conditions in the official method therefore define the result, and moisture values obtained by different methods are not interchangeable. Total solids is 100 minus moisture per cent.
Karl Fischer (KF) titration measures water specifically by its reaction with iodine and sulphur dioxide in the presence of a base and an alcohol. In volumetric KF, a reagent of known water equivalence is added until all water has reacted, detected electrometrically; in coulometric KF, iodine is generated electrochemically, suitable for very low water contents. KF is the reference method for low-moisture and sugar-rich products — dried foods, powders, chocolate, oils and fats, honey — and is not affected by other volatiles.
Distillation methods (such as the Dean–Stark method) co-distil water from the sample with an immiscible solvent such as toluene and measure the volume of water collected in a graduated trap; they suit spices and other products rich in volatile oils, where oven drying would overestimate moisture. Rapid methods for process control include infrared and halogen moisture balances (drying by radiant heat while weighing continuously), near-infrared spectroscopy calibrated against reference methods, microwave and capacitance sensors, and nuclear magnetic resonance. Water activity, distinct from moisture content (food-chemistry-major-constituents), is measured with dew-point or capacitance hygrometers.
pH is measured with a glass electrode and reference electrode connected to a pH meter, calibrated with at least two buffer solutions bracketing the expected value (commonly pH 4.0 and 7.0 for acid foods), with temperature compensation. pH governs microbial safety (the 4.6 boundary for low-acid foods, thermal-processing-canning), flavour, colour and the functionality of proteins and gels. Titratable acidity (TA) measures the total acid content by titrating with standard sodium hydroxide to a phenolphthalein end-point (about pH 8.2) or a set pH; it is expressed as the predominant acid: citric for most fruits, malic for apples, tartaric for grapes, lactic for dairy products and acetic for vinegar. TA (%) = (V × N × milliequivalent weight × 100)/sample mass, where the milliequivalent weight of citric acid (anhydrous) is 0.064 g, lactic acid 0.090 g, malic 0.067 g, tartaric 0.075 g and acetic 0.060 g. For example, 10 g of juice requiring 12.0 mL of 0.1 N NaOH contains 12.0 × 0.1 × 0.064 × 100/10 = 0.768 per cent acid as citric. pH and TA measure different things: pH reflects hydrogen-ion activity, which depends on buffering, while TA measures total titratable acid.
Refractometers measure the refractive index of a liquid, which rises with dissolved solids; calibrated in degrees Brix, they give the sucrose-equivalent soluble solids of juices, jams, syrups and honey, with temperature correction. The Brix/acid ratio is a standard maturity and quality index for fruit and juices. Colour, density and other physical tests are summarised in food-physical-properties-rheology.
Ash is the inorganic residue left after the organic matter of a food is burned off; it estimates total mineral content. In dry ashing, a weighed sample in a crucible is charred and then heated in a muffle furnace, commonly at about 525 to 550 °C, until a white or grey ash of constant weight remains: ash (%) = mass of ash/mass of sample × 100. Dry ashing is simple and handles many samples, but volatile elements such as mercury, arsenic, lead and selenium can be lost, and the ash may fuse with high-salt or high-phosphate samples. Wet ashing (wet digestion) oxidises organic matter with strong acids — typically nitric acid, sometimes with sulphuric acid, hydrogen peroxide or perchloric acid — at lower temperatures, retaining volatile elements, and is preferred as preparation for trace-element analysis; closed-vessel microwave digestion is now standard because it is fast, retains volatiles and reduces contamination.
Acid-insoluble ash, the part of the ash insoluble in dilute hydrochloric acid, consists mainly of silica and is a measure of sand, soil and dirt in spices, tea, flour and other products; standards set maximum limits for it, and it is a classic test for adulteration with earthy matter. Ash content is also used as an index of flour extraction rate (more bran gives more ash) and of fruit content in some products.
Some minerals are determined by classical chemical methods. Chloride, and hence salt content, is determined by titration with silver nitrate: in Mohr's method, in neutral solution with potassium chromate as indicator, the end-point is the first permanent red-brown silver chromate; in Volhard's method, excess silver nitrate is added and back-titrated with thiocyanate using ferric iron as indicator, suitable for acidic solutions. Salt is calculated from chloride by multiplying by 58.44/35.45 ≈ 1.65. For example, a sample containing 0.60 per cent chloride contains about 0.99 per cent sodium chloride. Calcium can be determined by EDTA complexometric titration, phosphorus by colorimetry as the molybdenum blue complex, and iron colorimetrically with reagents such as 1,10-phenanthroline. Ion-selective electrodes measure ions such as sodium, fluoride and nitrate directly in solution. Classical methods remain valuable because they need little equipment, are well understood and are often the official methods for compliance with product standards; their limitations are lower sensitivity and selectivity than instrumental methods and the skill needed for consistent end-points.
Atomic absorption spectroscopy (AAS) measures the absorption of light at an element-specific wavelength, produced by a hollow-cathode lamp of that element, by free atoms of the element in a flame or a graphite furnace. The absorbance follows the Beer–Lambert relation with concentration and is calibrated with standards. Flame AAS is robust and suitable for elements at milligram-per-kilogram levels, such as sodium, potassium, calcium, magnesium, iron and zinc; graphite-furnace (electrothermal) AAS is far more sensitive and is used for trace toxic elements such as lead and cadmium; hydride generation and cold-vapour techniques are used for arsenic, selenium and mercury. AAS measures one element at a time.
Inductively coupled plasma (ICP) techniques use an argon plasma at several thousand kelvin to atomise and ionise the sample. ICP optical emission spectrometry (ICP-OES) measures light emitted by excited atoms and ions at characteristic wavelengths and determines many elements simultaneously at milligram-per-kilogram levels. ICP mass spectrometry (ICP-MS) measures the ions directly by mass and has very low detection limits (micrograms per kilogram or lower) for trace and toxic elements and isotopes, making it the main technique for heavy-metal monitoring and for isotopic studies in authenticity testing. Spectral and mass interferences are corrected by choice of lines or isotopes, collision/reaction cells and internal standards.
The Kjeldahl method, developed in 1883, remains a reference method for protein in most foods. It has three stages. In digestion, the sample is heated with concentrated sulphuric acid, with potassium sulphate to raise the boiling point and a catalyst (copper sulphate is now usual; mercury and selenium catalysts are avoided for safety), converting organic nitrogen to ammonium sulphate. In distillation, the digest is made alkaline with sodium hydroxide, releasing ammonia, which is steam-distilled into a receiving solution of boric acid (or a known excess of standard acid). In titration, the ammonium borate formed is titrated with standard hydrochloric or sulphuric acid using a mixed indicator.
The nitrogen content is calculated as N (%) = (V_sample − V_blank) × N_acid × 14.007 × 100/(sample mass in mg), or equivalently (V_s − V_b) × N × 1.4007/sample mass in g, where volumes are in mL. For example, a 1.000 g sample requiring 15.20 mL of 0.1000 N HCl, with a blank of 0.20 mL, contains (15.00 × 0.1 × 1.4007)/1.000 = 2.10 per cent nitrogen, or 13.1 per cent protein using the factor 6.25. The blank corrects for nitrogen in the reagents.
Protein is calculated by multiplying nitrogen by a conversion factor. The general factor 6.25 assumes that protein contains 16 per cent nitrogen. Specific factors reflect the actual nitrogen content of particular proteins: commonly about 6.38 for milk and dairy products, 5.7 for wheat and many cereals, 5.95 for rice, 5.46 for peanuts, and others in tables of official methods; regulations and labelling rules specify which factor to use for each product, so the same nitrogen can give different declared protein values in different contexts.
The Dumas (combustion) method burns the sample at high temperature (around 900 to 1,000 °C or higher) in oxygen; nitrogen compounds are converted to nitrogen oxides, which are reduced to nitrogen gas, measured by a thermal conductivity detector after removal of water and carbon dioxide. Dumas analysis is fast (minutes per sample), automated and free of hazardous reagents, and has largely replaced Kjeldahl in many laboratories; because it measures nitrate and nitrite nitrogen as well, it gives slightly higher results than Kjeldahl for some materials. Both methods measure total nitrogen, including non-protein nitrogen such as urea, free amino acids, nucleic acids and nitrogenous adulterants like melamine, and both are therefore 'crude protein' methods.
Several colorimetric and spectroscopic methods measure protein more directly but are used mainly for specific purposes or in research. The biuret method measures the purple complex formed between peptide bonds and copper ions in alkaline solution; it is simple and specific for peptide bonds but insensitive. The Lowry method adds the Folin–Ciocalteu reagent to the biuret reaction for greater sensitivity, but many substances interfere. The bicinchoninic acid (BCA) method relies on copper reduction by protein and is sensitive and tolerant of some detergents. The Bradford (dye-binding) method uses Coomassie Brilliant Blue G-250, whose absorbance maximum shifts on binding to protein; it is rapid and sensitive but its response varies between proteins. Direct UV absorbance at 280 nm, due mainly to tryptophan and tyrosine, is used for purified proteins. Dye-binding methods such as amido black are standard in some countries for milk protein, and infrared spectroscopy is used for rapid routine analysis of milk and grain after calibration against Kjeldahl or Dumas. For nutritional quality rather than quantity, amino acid analysis by ion-exchange or reversed-phase chromatography after hydrolysis is used to calculate protein digestibility-corrected amino acid scores.
Fat (lipid) content is determined by extracting lipids with organic solvents and weighing the extract, or by volumetric methods in which fat is released and measured. In the Soxhlet method, a dried, ground sample in a porous thimble is extracted repeatedly with a solvent such as petroleum ether or diethyl ether, which is boiled, condensed and siphoned back through the sample over several hours; the solvent is evaporated and the fat weighed. Soxhlet measures 'free' fat that the solvent can reach in a dry sample. For example, 5.000 g of ground sample yielding 0.420 g of extract contains 8.40 per cent fat. Faster semi-automated extractors based on the same principle (such as Randall or hot-extraction systems) reduce extraction times, and accelerated solvent extraction uses elevated temperature and pressure.
Many foods contain fat bound to proteins and carbohydrates or enclosed in structures that solvent alone cannot reach — baked goods, cheese, meat products, eggs and many processed foods. For these, an acid hydrolysis step (the Weibull–Berntrop or acid-hydrolysis methods) is used before extraction, releasing bound lipids and giving 'total fat', which is the basis of nutrition labelling in many countries. For nutrition labelling, total fat is also defined in some jurisdictions as the sum of fatty acids expressed as triglycerides, determined by gas chromatography of fatty acid methyl esters (FAMEs), which also gives saturated, monounsaturated, polyunsaturated and trans fatty acids.
The Röse–Gottlieb and Mojonnier methods are reference methods for fat in milk and dairy products. The sample is treated with ammonia (to dissolve the casein and release fat) and ethanol, and then extracted with diethyl ether and petroleum ether in a special flask (the Mojonnier flask, which allows the ether layer to be decanted); the solvent layers are combined, evaporated and the fat weighed. These gravimetric methods are accurate and are used to calibrate rapid methods.
The Gerber method, widely used for routine milk testing, adds concentrated sulphuric acid to milk in a special calibrated glass butyrometer to dissolve proteins and release fat, with a small amount of amyl alcohol to help separate the fat; after centrifuging and warming in a water bath, the fat column is read directly on the butyrometer scale as per cent fat. The Babcock method, used mainly in North America, works on the same principle with a different bottle and without amyl alcohol. Both are quick and inexpensive; in large laboratories, mid-infrared milk analysers, calibrated against reference methods, measure fat, protein, lactose and total solids in seconds.
Beyond total fat, oils and fats are characterised by chemical constants and composition. The acid value (free fatty acids), peroxide value (hydroperoxides), iodine value (unsaturation), saponification value (mean molecular weight) and p-anisidine value (aldehydes) are explained in food-chemistry-major-constituents and are determined by titration or photometry. Fatty acid composition is determined by gas chromatography after converting triacylglycerols to fatty acid methyl esters by transesterification; polar capillary columns separate cis and trans isomers, giving trans fat content for labelling. Sterol composition, triacylglycerol profiles by high-performance liquid chromatography, and the Reichert–Meissl and Polenske values (volatile fatty acids) are used to detect adulteration of olive oil, butter and ghee with cheaper fats. Rapid methods for fat in meat, cereals and snacks include near-infrared spectroscopy and nuclear magnetic resonance. Choosing between gravimetric, volumetric and spectroscopic fat methods depends on the matrix, the required accuracy, the number of samples and whether the result is for labelling, for a product standard or for process control.
In many food composition tables and nutrition labels, carbohydrate is calculated by difference: carbohydrate (%) = 100 − (moisture + protein + fat + ash), sometimes also subtracting dietary fibre to give 'available' carbohydrate. This is simple but carries all the errors of the other determinations and includes non-carbohydrate components such as organic acids. Direct methods are preferred where accuracy matters. The phenol–sulphuric acid method measures total carbohydrate colorimetrically: concentrated sulphuric acid hydrolyses and dehydrates carbohydrates to furfural derivatives, which react with phenol to give a coloured product read at about 490 nm. The anthrone method works similarly. For energy calculation on labels, many jurisdictions now prefer available carbohydrate determined directly or by difference after subtracting fibre, and they set conversion factors (commonly 4 kcal/g for available carbohydrate and 2 kcal/g for fibre) that must be applied consistently. The choice of method therefore affects both the declared carbohydrate and the declared energy.
Reducing sugars are measured by their ability to reduce copper(II) or other oxidising agents. In the Lane–Eynon method, a standardised Fehling's solution (copper sulphate with alkaline tartrate) is titrated with the sugar solution at boiling, using methylene blue as an internal indicator that is decolorised when all copper has been reduced; results are calculated from standard tables or calibration with invert sugar or glucose. Sucrose, a non-reducing sugar, is measured by determining reducing sugars before and after acid or enzyme inversion: sucrose = (reducing sugars after inversion − reducing sugars before) × 0.95, the factor 0.95 (342/360) converting invert sugar to sucrose. The dinitrosalicylic acid (DNS) and Somogyi–Nelson methods are colorimetric alternatives.
Polarimetry measures the optical rotation of sugar solutions and is the classical method for sucrose in sugar factories (the sucrose content of raw sugar is expressed as 'pol'). Refractometry gives total soluble solids as °Brix. Enzymatic methods measure individual sugars specifically — glucose with glucose oxidase–peroxidase or hexokinase, lactose and sucrose after specific hydrolysis — and are widely available as kits. High-performance liquid chromatography with refractive-index or evaporative light-scattering detection, or ion chromatography with pulsed amperometric detection, separates and quantifies individual sugars (glucose, fructose, sucrose, maltose, lactose) in one run and is the reference approach for sugar labelling.
Starch is determined by enzymatic hydrolysis to glucose — with thermostable α-amylase and amyloglucosidase — followed by glucose measurement, after removing free sugars; polarimetric methods are used for some cereals. Resistant starch has its own enzymatic methods that mimic digestion.
Dietary fibre methods have evolved with its definition (food-chemistry-major-constituents). The older crude fibre method, boiling successively with dilute acid and alkali and weighing the residue, recovers only part of the cellulose and lignin and greatly underestimates fibre; it survives in some feed and legal standards. Detergent fibre methods (neutral and acid detergent fibre) are used mainly for feeds. For foods, the enzymatic–gravimetric method of Prosky (AOAC 985.29 for total dietary fibre, with related methods for soluble and insoluble fibre) digests the sample with heat-stable α-amylase, protease and amyloglucosidase to remove starch and protein, precipitates soluble fibre with ethanol, and weighs the residue, correcting for residual protein and ash. Newer integrated methods (such as AOAC 2009.01 and 2011.25) also capture resistant starch and the low-molecular-weight soluble fibres (such as inulin and resistant oligosaccharides) included in the Codex definition, giving higher and more complete values. Fibre declared on a label therefore depends on which method the regulator specifies.
Vitamins are present in foods at low concentrations (milligrams or micrograms per 100 g), often in several chemical forms with different activities, bound to proteins or other components, and many are sensitive to light, heat, oxygen, pH and metal ions. Vitamin analysis therefore requires careful sample handling — subdued or amber light, low temperatures, antioxidants such as ascorbic acid or BHT, inert atmospheres — and extraction steps that free the vitamins without destroying them: acid and enzyme hydrolysis (for example with takadiastase and papain) for B vitamins, alkaline saponification to remove fat and release vitamins A, D and E, and stabilising extractants such as metaphosphoric acid for vitamin C. Results are expressed in mass units or in activity units (such as retinol activity equivalents for vitamin A, combining retinol and provitamin A carotenoids). Because vitamin contents vary widely with variety, season, processing and storage, sampling for food composition tables follows careful designs, and analysed values are reported with the method used so that users can judge their comparability.
Vitamin C (ascorbic acid) is classically determined by titration with 2,6-dichlorophenolindophenol (DCPIP) dye, which ascorbic acid reduces from blue (pink in acid) to colourless; the end-point is the first persistent pink colour. The method is quick but measures only reduced ascorbic acid and is affected by other reducing substances and by coloured extracts; HPLC methods, often with reduction of dehydroascorbic acid to measure total vitamin C, are more specific. Thiamin (vitamin B1) is measured fluorometrically by the thiochrome method, in which thiamin is oxidised with alkaline ferricyanide to fluorescent thiochrome; riboflavin (B2) by its natural fluorescence. Fat-soluble vitamins — retinol, carotenoids, tocopherols and vitamin D — are measured by HPLC with UV, fluorescence or diode-array detection after saponification and extraction; vitamin D, present at very low levels, is increasingly measured by liquid chromatography–tandem mass spectrometry (LC-MS/MS).
HPLC has become the main technique for most vitamins because it separates the different forms (vitamers) and is specific; LC-MS/MS gives the greatest specificity and sensitivity and allows isotope-labelled internal standards.
Several B vitamins — folate, vitamin B12, biotin, pantothenic acid and niacin — have traditionally been measured by microbiological assays, in which the growth of a microorganism that requires the vitamin (for example Lactobacillus rhamnosus for folate and Lactobacillus delbrueckii subsp. lactis for vitamin B12) is proportional to the amount of vitamin in the sample, compared with standards. Microbiological assays measure total biologically active forms and are sensitive, but they are slow, need careful control of conditions and may respond to inactive analogues. They remain official methods for some vitamins and for fortified-food compliance, alongside newer LC-MS/MS and immunoassay methods (such as for vitamin B12 and folic acid in fortified foods). Vitamin analysis is central to verifying fortification of staples such as flour, rice, salt, oil and milk, a public-health measure in many countries. Laboratories verifying fortification must choose methods that measure the added form — for example folic acid rather than natural food folates — and must account for losses during processing and storage, which can be substantial for some vitamins.
Spectroscopy measures the interaction of electromagnetic radiation with matter. In ultraviolet–visible (UV-Vis) absorption spectrophotometry, light of a selected wavelength (about 190 to 400 nm in the UV and 400 to 800 nm in the visible) passes through a sample in a cuvette, and the fraction absorbed is measured. The Beer–Lambert law states that absorbance A = log(I0/I) = ε b c, where ε is the molar absorptivity (L mol⁻¹ cm⁻¹), b the path length (cm) and c the concentration (mol/L). Absorbance is therefore proportional to concentration within the linear range, usually up to an absorbance of about 1 to 1.5. For example, a compound with ε = 15,000 L mol⁻¹ cm⁻¹ giving an absorbance of 0.45 in a 1 cm cell has c = 0.45/15,000 = 3.0 × 10⁻⁵ mol/L. In practice, concentrations are read from a calibration curve of standards rather than from ε alone.
UV-Vis methods are used for coloured compounds and for colourless analytes after reaction with a reagent to form a coloured product: phosphorus as molybdenum blue, iron with phenanthroline, proteins by biuret or Bradford, total phenolics by the Folin–Ciocalteu method, anthocyanins by the pH-differential method, curcumin in turmeric, and many enzyme assays that follow NADH at 340 nm. A spectrophotometer contains a light source (deuterium lamp for UV, tungsten for visible), a monochromator, the sample cell and a detector; diode-array instruments record whole spectra at once.
In fluorescence, molecules absorb light at one wavelength (excitation) and emit it at a longer wavelength (emission) as they return to the ground state. Because emitted light is measured against a dark background rather than as a small decrease in a bright beam, fluorescence is typically much more sensitive than absorption — often by two to three orders of magnitude — and more selective, because both excitation and emission wavelengths are chosen. Fluorescence is used for riboflavin, thiamin (as thiochrome), vitamin B6, aflatoxins (which fluoresce naturally, the basis of their HPLC detection and of UV screening of kernels), some amino acids after derivatisation, and in many immunoassay and PCR detection systems. Quenching by other components and the inner-filter effect at high concentrations limit linearity and must be controlled. Instruments use a xenon lamp, excitation and emission monochromators or filters, and a detector at right angles to the beam to minimise scattered light.
Infrared (IR) spectroscopy measures absorption by molecular vibrations. In the mid-infrared region (about 4,000 to 400 cm⁻¹), absorption bands correspond to specific functional groups — O–H and N–H stretches near 3,300 cm⁻¹, C–H near 2,900 cm⁻¹, C=O (ester carbonyl of fats) near 1,745 cm⁻¹, amide I of proteins near 1,650 cm⁻¹, and isolated trans double bonds near 966 cm⁻¹, the basis of an official method for trans fat. Fourier-transform IR (FTIR) instruments with attenuated total reflectance (ATR) accessories allow direct measurement of liquids and solids with little preparation, and mid-IR milk analysers measure fat, protein and lactose.
Near-infrared (NIR) spectroscopy (about 780 to 2,500 nm) measures overtones and combinations of these vibrations; its bands are broad and overlapping, so it relies on multivariate calibration (chemometrics, such as partial least squares regression) against reference methods. Once calibrated, NIR measures moisture, protein, fat, starch and fibre in grains, flour, meat, dairy and feed in seconds without chemicals, and is used on-line and in grain trading. Calibrations must be developed and validated for each product and maintained with reference samples. Raman spectroscopy, measuring inelastic scattering of laser light, gives complementary vibrational information, is little affected by water and is used in authenticity testing and through-package analysis.
Chromatography separates the components of a mixture by their different distribution between a stationary phase and a mobile phase that moves past it. Components that interact more strongly with the stationary phase move more slowly and are separated from those that interact less. In thin-layer chromatography (TLC), the stationary phase is a thin layer of adsorbent such as silica gel on a plate; samples are spotted near the bottom, and a solvent mobile phase rises by capillary action. Each compound is characterised by its retardation factor, Rf = distance moved by the spot/distance moved by the solvent front; for example, a spot that moves 3.6 cm when the solvent front moves 8.0 cm has Rf = 0.45. TLC is cheap and fast and is used for screening — aflatoxins, synthetic colours, sugars, lipids and adulterants — and high-performance TLC with densitometry gives quantitative results. TLC is also widely used to check purity of standards and to choose solvent systems before developing column methods.
In gas chromatography (GC), the mobile phase is an inert carrier gas (helium, hydrogen or nitrogen) and the stationary phase a liquid film coating the inside of a long, narrow capillary column held in a temperature-programmed oven. The sample, which must be volatile or made volatile by derivatisation, is vaporised in a heated injector and separated by differences in boiling point and polarity. Detectors include the flame ionisation detector (FID), responding to almost all organic compounds and used for fatty acid methyl esters, alcohols and volatiles; the electron capture detector (ECD), very sensitive to halogenated compounds such as organochlorine pesticides; nitrogen–phosphorus and flame photometric detectors for organophosphorus pesticides; and mass spectrometers (GC-MS), which identify compounds. GC is used for fatty acid composition, trans fats, aroma and flavour compounds (often with headspace or solid-phase microextraction sampling), residual solvents, pesticide residues and ethanol and methanol in beverages. Quantification uses peak areas with internal standards added before extraction, which correct for injection variability and losses during sample preparation.
In high-performance liquid chromatography (HPLC), a liquid mobile phase is pumped at high pressure through a column packed with small particles. Most food applications use reversed-phase HPLC, with a non-polar stationary phase (commonly C18, octadecylsilane) and a polar mobile phase such as water–methanol or water–acetonitrile mixtures, often with a gradient of increasing organic solvent; polar compounds elute first. Normal-phase, ion-exchange, size-exclusion and hydrophilic-interaction modes are used for particular analytes. Detectors include UV and diode-array (most common), fluorescence (for aflatoxins, vitamins and derivatised amino acids), refractive index and evaporative light scattering (for sugars and lipids without chromophores), electrochemical and mass spectrometric detectors. HPLC is used for sugars, organic acids, vitamins, synthetic colours, preservatives, sweeteners, mycotoxins, phenolics and amino acids; ultra-high-performance LC (UHPLC) with sub-2 µm particles gives faster, sharper separations.
Separation quality is described by retention time (identification), resolution and efficiency. Resolution between two peaks is Rs = 2(t2 − t1)/(w1 + w2), where t are retention times and w baseline peak widths; Rs of 1.5 or more indicates baseline separation. For example, peaks at 6.2 and 7.0 min with widths of 0.40 and 0.40 min have Rs = 2 × 0.8/0.8 = 2.0. Column efficiency is expressed as the number of theoretical plates. Quantification uses peak area with external calibration or internal standards.
Mass spectrometry (MS) measures the mass-to-charge ratio (m/z) of ions. A mass spectrometer has an ion source, a mass analyser and a detector. Ionisation methods suited to GC include electron ionisation (EI), which fragments molecules in reproducible patterns that can be matched against spectral libraries, and chemical ionisation, which is gentler. For LC, the main techniques are electrospray ionisation (ESI), which produces ions from solution and suits polar and large molecules including peptides and proteins, and atmospheric-pressure chemical ionisation (APCI), for less polar compounds. Mass analysers include the quadrupole (filters ions by m/z with oscillating electric fields; robust and common), the ion trap, time-of-flight (TOF, measuring flight time; high mass accuracy and fast), and Orbitrap and Fourier-transform ion cyclotron resonance analysers with very high resolution and mass accuracy, which allow the elemental formula of an unknown to be determined. Mass spectrometers operate under high vacuum, and the interface between the chromatograph and the mass spectrometer, where solvent or carrier gas is removed, is a key part of instrument design.
Tandem mass spectrometry (MS/MS) selects a precursor ion in a first analyser, fragments it by collision with gas in a collision cell, and analyses the product ions in a second analyser. In triple-quadrupole instruments operated in multiple reaction monitoring (MRM) mode, specific precursor-to-product transitions are monitored for each analyte, giving outstanding selectivity and sensitivity in complex food matrices. LC-MS/MS and GC-MS/MS are the standard techniques for multi-residue analysis of pesticides and veterinary drugs, mycotoxins, process contaminants such as acrylamide, marine toxins, allergens (by signature peptides) and vitamin D. Regulatory confirmation of residues requires criteria such as retention-time agreement and ion-ratio tolerances for two or more transitions.
Accurate quantification in MS is complicated by matrix effects — suppression or enhancement of ionisation by co-eluting matrix components. They are corrected by matrix-matched calibration, standard addition, or best of all by stable-isotope-labelled internal standards (isotope dilution), which behave like the analyte but are distinguished by mass.
Hyphenated techniques couple a separation with a spectroscopic detector: GC-MS, LC-MS, LC-MS/MS, GC-olfactometry (in which a trained assessor sniffs the column effluent alongside a detector, to find odour-active compounds), LC-NMR and ICP-MS coupled to chromatography for element speciation (for example separating toxic inorganic arsenic from less toxic organic arsenic compounds in rice and seafood). High-resolution MS enables non-targeted analysis — screening for unknown contaminants and recording 'fingerprints' of foods — and is used in foodomics and authenticity testing (advanced-analytical-omics). Isotope-ratio mass spectrometry (IRMS), measuring tiny differences in ratios of stable isotopes such as ¹³C/¹²C, ¹⁵N/¹⁴N and ¹⁸O/¹⁶O, detects addition of cane or corn sugar to honey and juices and verifies geographical origin (food-fraud-authenticity). These techniques are expensive and need skilled staff, so they are concentrated in reference, research and large commercial laboratories, while routine plant laboratories rely on simpler methods and send confirmatory work out.
Electrophoresis separates charged molecules, chiefly proteins and nucleic acids, by their migration in an electric field through a gel. In sodium dodecyl sulphate–polyacrylamide gel electrophoresis (SDS-PAGE), proteins are denatured and coated with the anionic detergent SDS, often after reduction of disulphide bonds with mercaptoethanol or dithiothreitol; SDS gives them a nearly uniform charge-to-mass ratio, so they separate according to molecular weight, smaller proteins moving faster through the gel. Molecular weights are estimated by comparison with a ladder of marker proteins, and bands are stained with Coomassie blue or silver. Native PAGE keeps proteins in their native state and separates by charge, size and shape. Isoelectric focusing separates proteins by isoelectric point in a pH gradient, and two-dimensional electrophoresis combines isoelectric focusing with SDS-PAGE to resolve hundreds of proteins, as in proteomics.
Food applications include identifying species in meat and fish products by their protein patterns, detecting cow's milk in sheep or goat cheese, characterising wheat gluten subunits for baking quality, following proteolysis during cheese ripening and meat ageing, and checking protein ingredients. Capillary electrophoresis, performed in narrow fused-silica capillaries with on-line detection, gives fast, automated separations of proteins, peptides, organic acids and ions. DNA-based methods (PCR) are now preferred for species identification in processed foods because DNA survives heating better than protein patterns.
Immunoassays use the specific binding between an antibody and its target (antigen). The enzyme-linked immunosorbent assay (ELISA) links this binding to an enzyme reaction that produces a measurable colour. In a sandwich (non-competitive) ELISA, used for large analytes with at least two binding sites such as allergenic proteins, the analyte is captured by an antibody fixed to the well and detected by a second, enzyme-labelled antibody; colour increases with analyte concentration. In a competitive ELISA, used for small molecules such as mycotoxins, drug residues and pesticides that have only one binding site, the analyte in the sample competes with a labelled analyte for a limited number of antibody sites; colour decreases as analyte concentration increases. ELISAs are sensitive, relatively cheap and suitable for many samples, and are the main methods for food allergens (gluten, peanut, milk, egg) and for screening mycotoxins and veterinary drug residues.
Limitations include cross-reactivity with related compounds, effects of processing on antibody recognition (heat can change epitopes, causing underestimation of allergens in cooked foods), matrix interference and the dependence of results on the antibody and calibrant used; results from different kits can differ. Positive screening results for regulated contaminants are therefore usually confirmed by chromatographic or mass-spectrometric methods.
Lateral flow devices (immunochromatographic strips), similar in principle to home pregnancy tests, carry antibodies on a membrane strip; the sample flows along the strip by capillary action, and a coloured line appears at a test zone depending on the presence of the analyte, with a control line confirming that the strip worked. They give results in minutes without instruments and are used for allergen swabs on cleaned equipment, rapid screening of mycotoxins in grain at intake, antibiotic residues in milk tankers and detection of pathogens after enrichment. Readers can make them semi-quantitative. Other rapid methods include biosensors, ATP bioluminescence for hygiene monitoring and PCR kits for species and pathogens. Rapid methods support decisions at intake and on the line, but their performance must be verified for each matrix, and important results are confirmed by reference methods. Records of rapid-test results, including the lot numbers of strips and kits, are kept as part of the food safety management system, because they show that intake and cleaning controls were actually carried out.
Method validation is the process of demonstrating, with objective evidence, that a method is fit for its intended purpose. The main performance characteristics are: selectivity (specificity) — the ability to measure the analyte without interference from other components; trueness (accuracy of the mean), assessed by analysing certified reference materials or by recovery of spiked analyte: recovery (%) = (measured in spiked sample − measured in unspiked sample)/amount added × 100 — for example, a sample containing 2.0 mg/kg spiked with 5.0 mg/kg that measures 6.6 mg/kg shows a recovery of (6.6 − 2.0)/5.0 = 92 per cent; precision — the closeness of agreement between repeated results, expressed as standard deviation or relative standard deviation (RSD, also called coefficient of variation), under repeatability conditions (same analyst, equipment, short time) and reproducibility conditions (different laboratories); linearity and working range of the calibration; and sensitivity, expressed as the slope of the calibration.
The limit of detection (LOD) is the lowest concentration that can be reliably distinguished from a blank, and the limit of quantification (LOQ) the lowest that can be measured with acceptable precision and trueness. Common estimates use the standard deviation of blank or low-level responses (σ) and the calibration slope (S): LOD ≈ 3.3σ/S and LOQ ≈ 10σ/S, or the mean blank plus 3 or 10 standard deviations. Ruggedness (robustness) testing checks how small deliberate changes in conditions affect results. For methods used to enforce limits, the LOQ must be well below the regulatory limit.
Every result has an uncertainty — a parameter characterising the range within which the true value is believed to lie. It combines contributions from sampling, weighing, volumetric operations, calibration, recovery and precision, estimated either by a component-by-component approach or from validation and quality-control data (top-down). Results are reported as value ± expanded uncertainty, usually with a coverage factor k = 2 for about 95 per cent confidence. Uncertainty matters when results are close to legal limits: regulators in many systems take a product to be non-compliant only when the result minus its expanded uncertainty exceeds the limit.
Statistical tools used in the laboratory include the mean, standard deviation and RSD; t-tests and F-tests to compare methods and analysts; outlier tests such as Grubbs' and Dixon's; regression for calibration; and for collaborative studies, the Horwitz equation, an empirical relation predicting that the reproducibility RSD increases as concentration decreases (RSD_R (%) ≈ 2C^(−0.1505), where C is the concentration as a mass fraction). The HorRat ratio — observed reproducibility RSD divided by the Horwitz-predicted value — is used to judge whether a method's precision is acceptable, values of about 0.5 to 2 being considered normal.
Internal quality control keeps a validated method in control day to day. Each analytical batch includes reagent blanks, calibration standards, replicate analyses, spiked samples to check recovery, and control samples or certified reference materials whose results are plotted on control charts (Shewhart charts with warning limits at ±2 standard deviations and action limits at ±3); rules such as the Westgard rules flag runs to be rejected. External quality assessment through proficiency testing — analysing samples distributed by a scheme provider and receiving a z-score comparing the laboratory's result with the assigned value (|z| ≤ 2 satisfactory, 2 < |z| < 3 questionable, |z| ≥ 3 unsatisfactory) — checks performance against other laboratories.
Laboratory quality management is formalised in ISO/IEC 17025, the international standard for the competence of testing and calibration laboratories, covering impartiality, personnel, equipment calibration and traceability, method validation, measurement uncertainty, quality control, reporting and management of non-conformities. Accreditation to ISO/IEC 17025 by national bodies — for example NABL in India, UKAS in the United Kingdom and A2LA and ANAB in the United States — is required for laboratories testing food for regulatory purposes in many countries, including laboratories notified under food safety law. Good laboratory practice (GLP) is a separate quality system for non-clinical safety studies. Traceability of measurements to SI units through calibrated balances, volumetric ware and certified reference materials underpins all of it.
A student uses this subject in every laboratory course: determining moisture, ash, protein by Kjeldahl, fat by Soxhlet and Gerber, reducing sugars by Lane–Eynon, vitamin C by DCPIP and acidity by titration, and calculating results correctly with blanks and factors. They learn to run a spectrophotometer calibration curve, a TLC plate and an HPLC sequence, and to report results with replicates. In examinations they answer questions on method principles, calculations and which technique suits which analyte. They also learn laboratory safety for acids, solvents and gases.
An analyst or quality-control chemist five years into a food plant or a regulatory laboratory runs proximate analyses and specific tests every day, maintains instruments, prepares calibration standards, plots control charts, takes part in proficiency testing, writes test reports, and decides whether results show compliance once measurement uncertainty is considered. In a regulatory laboratory, the same person follows legal sampling and chain-of-custody procedures and may defend results as an expert witness. They also train plant staff in sampling and rapid tests, and investigate out-of-specification results before product is released.
A laboratory manager or head of quality with long experience designs the laboratory's scope, chooses instruments and methods, leads ISO/IEC 17025 accreditation and audits, validates new methods for new regulations (new contaminant limits, labelling rules), manages a network of plant and contract laboratories, and interprets results for management and regulators when a product is near or over a limit or a dispute arises with a supplier or authority. They also manage budgets, staffing and the balance between in-house and outsourced testing.
A research leader develops new analytical methods — non-targeted high-resolution MS, rapid spectroscopic and sensor methods, DNA and proteomic authenticity tests — leads collaborative studies that turn methods into official standards, advises regulators on performance criteria and limits of quantification for new contaminants, contributes to bodies such as AOAC, ISO and Codex committees on methods of analysis, and trains analysts and examiners. They also judge when emerging methods are mature enough to be adopted for official control, and how to interpret results near legal limits when methods disagree.
This manual summarises principles, calculations and applications; it does not reproduce official method texts, whose current versions must be followed in practice. Legal sampling procedures are described in general terms because the number of parts, sealing and dispatch rules differ by country and law. Microbiological methods are in food-microbiology; DNA-based authenticity methods in food-fraud-authenticity; non-targeted omics in advanced-analytical-omics. Specific conversion factors, tolerance limits and acceptance criteria quoted are common textbook values; regulations and official methods specify the values that apply.
Original practice questions written to the syllabus and pattern of the exams named in each question's examTags. They are not past papers and do not reproduce any board's questions. Every numerical answer was computed by script. Items tagged to food analyst, food safety officer and other licensing or recruitment exams whose syllabi Laisha has not yet verified are written to the expected scope of those exams.
These notes are free to read here, with no account.
What the Laisha app adds:
Free for ten days, then a subscription. We would rather write that here than have you find it later.