Veterinary · Para-clinical · also MSVE LPT-311, Dairy Technology
Milk composition and quality testing, adulteration, heat processing, indigenous and fermented dairy products, dairy standards, packaging and quality control.
In the exams: ICAR AICE-JRF core group 05; ICAR AIEEA-PG 14 (LPT); TNPSC LPT 15; RPSC LPT (BIS, HACCP)
Milk and milk products technology is the science of turning raw milk into safe, standard, stable foods. It starts at the farm gate, where milk is tested and accepted or rejected, and runs through the processing hall (clarification, standardisation, pasteurisation, homogenisation, UHT treatment), the product sections (cream, butter, ghee, paneer, khoa, dahi, cheese, ice cream, condensed and dried milks) and the packing line, and ends with the hygiene, quality-control and legal standards that let a product reach the consumer.
The subject matters to a veterinarian for two reasons. First, the dairy plant is where animal health becomes public health: mastitis milk, antibiotic residues, aflatoxin M1 and milk-borne zoonoses are all caught, or missed, at the reception dock and in the pasteuriser. Second, veterinarians run or advise milk cooperatives, chilling centres and plants, and need the arithmetic of standardisation, yield and pricing.
The manual is global but keeps the traditional South Asian products in full, because ghee, paneer, khoa, chhena and dahi carry most of the milk produced in India, Pakistan, Bangladesh and Nepal, and because Indian examinations ask about them in detail. Compositional figures are typical values; real milk varies with species, breed, stage of lactation, season and feed, and legal standards vary by country and, in India, by state. Always confirm a legal limit against the current text of the regulation in force.
In India the subject is examined in ICAR AIEEA-PG (Animal Sciences, LPT), the AICE-JRF/SRF core group 05, the ASRB NET for LPT, and state veterinary officer papers (TNPSC has a 15-question LPT block; RPSC names BIS and HACCP).
Milk is the secretion of the mammary gland. Chemically it is an emulsion of fat globules and a colloidal suspension of casein micelles in a true solution of lactose, whey proteins, minerals and vitamins. Its constituents are grouped as water, fat and solids-not-fat (SNF); SNF is everything other than water and fat, mainly protein, lactose and ash. Total solids (TS) are fat plus SNF.
Typical cow milk is about 87 per cent water, 3.5 to 4.5 per cent fat, 3.2 to 3.5 per cent protein, 4.6 to 4.9 per cent lactose and 0.7 per cent ash, giving an SNF of about 8.5 to 9.0 per cent. Buffalo milk is richer: fat commonly 6 to 8 per cent (range wider), protein about 4 to 4.5 per cent and SNF about 9.5 to 10 per cent. Goat milk is close to cow milk in gross composition but has smaller fat globules and more short- and medium-chain fatty acids (caproic, caprylic, capric). These figures are typical, not fixed; they shift with breed, stage of lactation, season, feeding and milking interval.
Buffalo and goat milk contain little or no beta-carotene because those species convert it to vitamin A, so their milk fat, butter and ghee are white; cow ghee is yellow. Buffalo milk also has larger fat globules, more casein and calcium, and a higher buffering capacity, which is why it gives firmer paneer, more khoa per litre and a different heat stability from cow milk.
Milk protein is about 80 per cent casein and 20 per cent whey (serum) protein in cow milk. The caseins (alpha-s1, alpha-s2, beta and kappa) form micelles held together by colloidal calcium phosphate; kappa-casein sits on the surface and keeps the micelle stable. Casein precipitates at its isoelectric pH of about 4.6 (the basis of dahi, chhena and paneer) and is also clotted by rennet, which cuts kappa-casein. The whey proteins, mainly beta-lactoglobulin and alpha-lactalbumin, stay soluble at pH 4.6 but are denatured by heat: beta-lactoglobulin unfolds above about 70 to 75 degrees C and releases sulphydryl groups that give cooked flavour.
Lactose is a disaccharide of glucose and galactose joined by a beta-1,4 bond. It is the substrate for lactic fermentation, crystallises easily (sandiness in ice cream and condensed milk), and reacts with protein in the Maillard reaction that browns heated milk. Milk fat is about 98 per cent triglycerides, carried in globules surrounded by a membrane of phospholipid and protein (the milk fat globule membrane). It is unique among common fats in containing butyric acid (C4:0), which is why the Reichert-Meissl value is used to detect adulteration of ghee.
Milk also carries native enzymes that are used as process indicators: alkaline phosphatase (destroyed by pasteurisation), lactoperoxidase (survives pasteurisation, destroyed at about 80 degrees C) and lipoprotein lipase (causes rancidity if the fat globule membrane is damaged).
Fresh milk has a pH of about 6.6 to 6.8 and a titratable acidity, expressed as lactic acid, of about 0.13 to 0.14 per cent for cow milk and slightly higher for buffalo milk. This fresh-milk acidity is called natural or apparent acidity; it comes from casein, phosphates, citrates and dissolved carbon dioxide, not from lactic acid. Acidity above this, produced by bacterial fermentation of lactose, is developed or real acidity.
The specific gravity of whole milk is about 1.028 to 1.032 at 15.5 degrees C (60 degrees F); fat lowers it and SNF raises it. Adding water lowers specific gravity; removing fat raises it, so a skilled adulterator can skim and water milk together and leave the lactometer reading unchanged. That is why lactometer readings are always interpreted alongside the fat test.
The lactometer reading (L) is the second and third decimal places of specific gravity multiplied by 1000 (specific gravity 1.030 is L = 30). A Quevenne lactometer is calibrated at 60 degrees F; by the convention taught in Indian laboratories, 0.1 is added to the reading for each degree F above 60 and subtracted for each degree below, giving the corrected lactometer reading (CLR). Richmond's formula then estimates solids: TS per cent = 0.25 L + 1.2 F + 0.14, so SNF per cent = 0.25 L + 0.2 F + 0.14, where F is fat per cent. National standards publish modified versions of this formula, so state which one you use.
The freezing point of authentic milk is very constant, close to minus 0.52 to minus 0.54 degrees C, because it is fixed by the osmotic pressure of blood. Added water raises it towards zero, which makes the cryoscope the reference method for added water.
Platform tests are quick tests done at the reception dock to accept or reject each can or tanker before milk enters the plant.
Organoleptic test: the tester smells and looks at milk in the can as the lid is lifted. Off-odours (sour, feed, cowy, rancid, chemical), colour change, foreign matter and abnormal consistency lead to rejection. It is the first and fastest screen.
Clot-on-boiling (COB) test: a small sample is boiled. Milk that clots has high developed acidity or is abnormal (colostrum, some mastitis milk) and will not withstand pasteurisation; it is rejected. The COB test is positive only when acidity is well above normal.
Alcohol test: equal volumes of milk and ethanol (68 per cent is common; some countries use 72 or 75 per cent) are mixed. Flakes show that the casein is unstable to heat because of developed acidity, colostrum, mastitis, or a calcium-magnesium imbalance. Because it is more sensitive than the COB test, it is used to select milk for heat-intensive processes such as UHT and condensing.
Titratable acidity: milk is titrated with 0.1 N sodium hydroxide to a phenolphthalein end point. Each millilitre of 0.1 N alkali neutralises 0.009 g of lactic acid, so acidity per cent = (mL of 0.1 N NaOH x 0.009 x 100) / weight of sample in grams.
Sediment test: a measured volume is filtered through a cotton disc, and the dirt left is compared with standard discs. It reflects the hygiene of milking.
The Gerber method is the routine reference method for fat in milk across most of the world. Into a butyrometer go 10 mL of sulphuric acid (specific gravity about 1.807 to 1.812), 10.75 mL of milk (delivered by the 11 mL pipette, because some milk clings to the pipette) and 1 mL of amyl alcohol. The acid dissolves the protein and the fat globule membrane, freeing the fat; amyl alcohol sharpens the fat column and prevents charring. The butyrometer is stoppered, shaken, centrifuged (around 1100 revolutions per minute for 4 to 5 minutes), held in a water bath at 65 degrees C and read directly as per cent fat.
Other methods include the Babcock test (used historically in North America), the Rose-Gottlieb solvent extraction method (the gravimetric reference method, used for products as well as milk), and infrared milk analysers and ultrasonic analysers that read fat, protein, lactose and SNF in a minute and are calibrated against the reference methods.
Methylene blue reduction test (MBRT): methylene blue is added to milk, which is held at 37 degrees C. Bacteria consume dissolved oxygen and reduce the dye to colourless; the more bacteria, the faster the colour disappears. A long reduction time means good milk. Indian standards grade raw milk by reduction time, with five hours or more as the best grade and half an hour or less as poor. The test measures metabolic activity, not number, and psychrotrophs in chilled milk reduce the dye slowly, so it underrates the count of cold-stored milk.
Resazurin test: resazurin changes from blue through mauve and pink to colourless as it is reduced, so the colour at a fixed time (10 minutes or 1 hour) can be compared with a disc.
Standard plate count (SPC) or total bacterial count: milk is plated on plate count agar and incubated (commonly 30 degrees C for 72 hours under ISO methods; other temperatures are used in some countries). It is the legal basis of raw milk grading in many countries. Coliform count (on violet red bile agar or by the most probable number method) indicates unhygienic production and, in pasteurised milk, post-pasteurisation contamination.
Somatic cell count (SCC) measures white cells and some epithelial cells; it rises with mastitis. The California Mastitis Test (CMT) is the cow-side screen: a detergent reagent breaks the cells and the released DNA forms a gel graded by eye. Legal bulk-milk limits differ: 400,000 cells per mL in the EU and 750,000 per mL for cow milk under the US Pasteurized Milk Ordinance, with different limits for goats.
Residue screening at reception uses microbial inhibitor tests (spores of a sensitive thermophilic bacillus that fail to grow and change colour if antibiotics are present) and rapid receptor or lateral-flow assays for specific drug classes.
Adulteration is the addition of anything to milk, or the removal of anything from it, that lowers its quality or misleads the buyer. Because milk is paid on volume, fat and SNF, the cheapest frauds are adding water and removing fat. Others mask those frauds or extend shelf life: sugar, starch, glucose, salt, urea and skim milk powder are added to restore lactometer reading or SNF; neutralisers such as sodium carbonate, bicarbonate or hydroxide are added to hide developed acidity; and preservatives such as formalin, hydrogen peroxide and boric acid are added to stop souring. So-called synthetic milk, a mixture of water, vegetable oil, detergent (to emulsify the oil), urea and sugar, has been reported in South Asia. Melamine added to raise apparent protein (nitrogen) caused deaths of infants in China in 2008 and led Codex to set maximum levels for melamine.
Adulteration with water lowers fat, SNF and specific gravity and raises the freezing point. Skimming lowers fat but raises specific gravity. The two together can leave the lactometer reading normal, so fat and SNF must both be checked.
Two calculations are standard. Added water, as per cent of the adulterated sample, can be estimated from SNF: (SNF of genuine milk minus SNF of sample) divided by SNF of genuine milk, times 100. By the freezing point method: (T minus T') divided by T, times 100, where T is the freezing point depression of genuine milk and T' that of the sample. Fat abstracted, as per cent of the original fat: (F minus F') divided by F, times 100.
Starch and cereal flours: add iodine solution to boiled and cooled milk; a blue colour shows starch.
Cane sugar (sucrose): the resorcinol test. Milk is heated with resorcinol in hydrochloric acid; a red colour shows sucrose, which reacts as a ketose. Lactose does not give the colour under the test conditions.
Urea: p-dimethylaminobenzaldehyde (DMAB) gives a distinct yellow colour with urea. Milk normally contains a little urea, so the test is read against a control.
Neutralisers (carbonate, bicarbonate, hydroxide): the rosalic acid test. Milk mixed with alcohol and rosalic acid turns rose-red if alkali has been added; pure milk gives a brownish colour. A raised ash alkalinity also points to added neutraliser.
Formalin: Hehner's test. Sulphuric acid containing a trace of ferric chloride is run down the side of a tube of milk; a violet or purple ring at the junction shows formaldehyde.
Hydrogen peroxide: p-phenylenediamine turns blue in the presence of peroxide (the vanadium pentoxide test is an alternative).
Boric acid and borax: the turmeric paper test. Acidified milk on turmeric paper, dried, gives a red colour that turns green-blue with ammonia.
Common salt: silver nitrate with potassium chromate indicator; excess chloride leaves the mixture yellow instead of brick-red. Commercial rapid kits for several adulterants at once are widely used by cooperatives and regulators in India; they screen, and a positive result is confirmed by a reference method.
The legal response is set by national law. In India, adulteration is dealt with under the Food Safety and Standards Act 2006 and its regulations, with sampling by food safety officers and analysis in notified laboratories.
At the reception dock milk is graded (Chapter 2), weighed or metered, sampled for payment, and pumped through a filter into a chilled silo. Raw milk should be cooled quickly after milking; most codes ask for 4 to 5 degrees C or below within a few hours, and bulk milk coolers at village collection centres in South Asia serve that purpose.
Clarification removes dirt, cells and sediment by centrifugal force in a clarifier, a disc-stack centrifuge in which the heavy particles are thrown outwards and collected as sludge.
Bactofugation, high-speed centrifugation, removes most bacterial spores and is used for cheese milk to prevent late blowing by Clostridium tyrobutyricum.
Separation splits whole milk into cream and skim milk in a separator, also a disc-stack centrifuge, using the lower density of fat. An efficient separator leaves about 0.05 to 0.1 per cent fat in skim milk.
Standardisation adjusts fat and, where required, SNF to a legal or commercial target. It can be done by mixing two milks or by adding cream, skim milk, skim milk powder or water.
The Pearson square solves a two-ingredient blend for one component. Write the target fat percentage in the centre, the fat percentages of the two ingredients on the left corners, and subtract diagonally (always the smaller from the larger). Each number on the right is the parts of the ingredient on the same horizontal line. For example, to make toned milk of 3.0 per cent fat from buffalo milk of 6.5 per cent and skim milk of 0.1 per cent: buffalo milk parts = 3.0 minus 0.1 = 2.9; skim milk parts = 6.5 minus 3.0 = 3.5; total 6.4 parts. For 1000 kg, buffalo milk = 1000 x 2.9 / 6.4 = 453.1 kg and skim milk = 546.9 kg.
When both fat and SNF must be met, use mass balance: write one equation for total mass, one for fat and one for SNF, and solve. To make 1000 kg of toned milk with 3.0 per cent fat and 8.5 per cent SNF from buffalo milk (6.8 per cent fat, 9.4 per cent SNF), skim milk powder (taken here as fat-free and 96 per cent SNF) and water: fat gives buffalo milk = 30 / 0.068 = 441.2 kg; this supplies 41.5 kg SNF, so the powder must supply 85 minus 41.5 = 43.5 kg SNF, which is 45.3 kg of powder; water = 1000 minus 441.2 minus 45.3 = 513.5 kg. Toned milk made this way, buffalo milk extended with reconstituted skim milk powder, was introduced in India in the 1950s to stretch buffalo milk for city supply; double toned milk carries the idea further.
Homogenisation breaks fat globules (average diameter about 3 to 4 micrometres in raw cow milk) into much smaller ones, mostly below 1 micrometre, by forcing warm milk through a narrow gap at high pressure. The globules are torn by turbulence and cavitation, and the new surface is coated with casein and whey protein. A typical two-stage homogeniser uses a high first-stage pressure followed by a much lower second stage; the second stage breaks up clusters of globules that form after the first.
Homogenised milk has no cream line, looks whiter, tastes richer and fuller, forms a softer curd in the stomach and suits recombined and flavoured milks, cream liqueurs and ice cream mix. Its drawbacks are greater sensitivity to light-induced off-flavour, lower heat and acid stability of the casein in some products, and a strong tendency to lipolysis if raw milk is homogenised and not heated at once, because native lipase attacks the new, unprotected fat surface. Milk is therefore homogenised warm (around 55 to 75 degrees C) and pasteurised immediately before or after.
Pasteurisation is a heat treatment designed to destroy all vegetative pathogens likely to be present in milk, with minimal change to its composition, flavour and nutritive value. It does not sterilise; spores and some heat-resistant bacteria survive, and pasteurised milk must be kept chilled.
The classic combinations are low-temperature long-time (LTLT, batch or holder pasteurisation) at 63 degrees C (145 degrees F) for 30 minutes, and high-temperature short-time (HTST, continuous) at 72 degrees C (161 degrees F) for 15 seconds. The standards were originally set to kill Mycobacterium tuberculosis and M. bovis; the target organism was later changed to Coxiella burnetii, the agent of Q fever, which is slightly more heat-resistant, and the 63 degrees C figure was raised from the earlier 61.7 degrees C for that reason. Products with more fat or sugar, such as cream and ice cream mix, receive higher temperatures or longer times because fat and solids protect bacteria; national rules give the exact values. Thermisation (about 57 to 68 degrees C for around 15 seconds) is a milder treatment of raw milk to reduce psychrotrophs before storage, and is not pasteurisation.
Alkaline phosphatase (ALP) is destroyed by a time-temperature combination close to pasteurisation, so a negative ALP test proves adequate heating, and a positive test shows under-heating or contamination with raw milk. Many regulators set a numerical ALP limit for pasteurised cow milk; the method and limit are given in national law. Lactoperoxidase survives pasteurisation and is destroyed at about 80 degrees C; a negative peroxidase test therefore indicates heating well above pasteurisation, which is required for some products and unwanted for others.
An HTST pasteuriser is built around a plate heat exchanger with sections for regeneration, heating, holding and cooling. Cold raw milk enters the regeneration section, where it is warmed by hot pasteurised milk flowing on the other side of the plates; the pasteurised milk is cooled at the same time, saving energy. The milk is then heated to pasteurisation temperature by hot water, passes through a holding tube sized to give at least 15 seconds, and returns through regeneration and finally through chilled water or glycol sections to about 4 degrees C.
Regeneration efficiency is the temperature rise achieved in regeneration divided by the total temperature rise from inlet to pasteurisation temperature, times 100. For raw milk entering at 5 degrees C, leaving regeneration at 60 degrees C and heated to 72 degrees C, efficiency = 55 / 67 x 100 = 82 per cent.
The holding tube length is fixed by flow rate and tube diameter. Hold time is the volume of the tube divided by the flow rate. Because flow in a tube is not uniform, the tube must be long enough for the fastest particle, not the average one; in fully laminar flow the fastest particle moves at twice the mean velocity, so regulators who design for the fastest particle demand about double the length that the mean velocity would suggest.
Safety devices are central. A timing pump (or flow controller) fixes the maximum flow. A recording thermometer at the end of the holding tube produces the legal record. A flow diversion valve sends milk back to the balance tank whenever its temperature falls below the set point. Pressure on the pasteurised side of the regeneration section must be higher than on the raw side, so that any pinhole leak carries pasteurised milk into raw milk and not the other way. Each of these is a critical control point in a dairy HACCP plan (Chapter 11).
Pasteurisation causes little change in nutrients; there is a small loss of heat-sensitive vitamins (vitamin C, thiamine, B12) and slight whey protein denaturation.
Shelf life of pasteurised milk depends far more on post-pasteurisation contamination and storage temperature than on the heat treatment itself. Recontamination from fillers and pipes introduces Gram-negative psychrotrophs that grow at refrigeration temperature, while heat-resistant spore formers such as Bacillus cereus survive pasteurisation and limit shelf life at the higher end of chill storage.
Ultra-high-temperature (UHT) processing heats milk continuously to about 135 to 150 degrees C for a few seconds, cools it, and fills it aseptically into sterile packages. The aim is commercial sterility: no microorganisms capable of growing under normal non-refrigerated storage. UHT milk keeps for several months at ambient temperature unopened.
Two heating systems are used. In indirect systems, heat passes through a wall: plate, tubular or scraped-surface heat exchangers. In direct systems, culinary steam is injected into milk (steam injection) or milk is sprayed into steam (steam infusion), and the condensed water is removed by flash cooling in a vacuum chamber that brings the milk back to its original solids. Direct heating is faster up and down, so it gives less cooked flavour and less chemical change for the same kill, at higher energy cost.
In-container sterilisation heats milk already sealed in bottles or cans, typically in the range of 110 to 120 degrees C for 10 to 20 minutes or more, in batch or continuous hydrostatic sterilisers. It gives more browning, caramel flavour and vitamin loss than UHT.
Heat lethality is described by the D-value (time at a given temperature to kill 90 per cent, one log, of a population) and the z-value (temperature change that alters D tenfold). The number of log reductions is the holding time divided by D. For low-acid canned foods the reference is the 12D process for Clostridium botulinum spores; at 121.1 degrees C a D of about 0.2 to 0.25 minute gives about 3 minutes, the minimum F0. For UHT, thermophilic spore formers such as Geobacillus stearothermophilus are more relevant to spoilage, and Bacillus sporothermodurans is a known survivor.
UHT milk can develop cooked or sulphurous flavour (from denatured beta-lactoglobulin), stale flavour on storage, Maillard browning, sediment, and age gelation, a slow thickening months after processing. Age gelation and bitterness are often due to heat-stable proteases from psychrotrophic bacteria (mainly Pseudomonas) that grew in the raw milk before processing: the bacteria die, but their enzymes survive UHT. Rancidity similarly comes from heat-stable bacterial lipases. Good raw milk (low psychrotroph count, short cold storage) is therefore the main control.
Indices of heat treatment help regulators tell products apart. Lactulose, formed from lactose by heat, is low in UHT and high in in-container sterilised milk. Undenatured whey protein (measured by the turbidity test of Aschaffenburg or by chromatography) is high in pasteurised milk, reduced in UHT and almost absent in sterilised milk; the turbidity test, in which ammonium sulphate precipitates casein and denatured whey protein and the filtrate is boiled, shows no turbidity in properly sterilised milk.
Evaporated milk is milk concentrated about two to two-and-a-half times under vacuum, homogenised, filled in cans and sterilised. Its heat stability during sterilisation is the main technical issue; it is improved by forewarming the milk before evaporation and by adding stabilising salts (phosphates or citrates) after a pilot sterilisation test.
Sweetened condensed milk is concentrated milk with sucrose and is preserved by the high osmotic pressure of the sugar, not by sterilisation. The key number is the sugar ratio, the sugar in the water phase: sugar divided by (sugar plus water), times 100, usually about 62.5 to 64.5 per cent. Below that range the product spoils; above it sucrose may crystallise. Lactose becomes supersaturated as the product cools; it is seeded with very fine lactose powder and cooled with stirring so that many tiny crystals form instead of a few large ones that would feel sandy.
Milk powders are made by evaporation to a concentrate (around 45 to 50 per cent solids) and then drying. Spray drying atomises the concentrate into hot air in a chamber; droplets dry in seconds at low particle temperature because evaporation cools them, giving a powder of good solubility. Roller (drum) drying spreads milk on steam-heated drums and scrapes off the film; it is cheaper but gives scorched particles and poorer solubility, and is now used for some feed and confectionery powders. Skim milk powder for different uses is classed by the whey protein nitrogen index (WPNI), which measures undenatured whey protein: low-heat powder for recombined milk and cheese, high-heat powder for bakery and recombined evaporated milk. Whole milk powder oxidises in air, so it is packed under nitrogen or nitrogen-carbon dioxide gas in cans or barrier pouches.
Cream is the fat-rich part of milk obtained by separation. Commercial classes run from light or table cream (around 18 to 25 per cent fat) through whipping cream (30 to 40 per cent) to heavy and plastic creams; the legal names and limits differ by country.
Butter is made by churning cream. Agitation damages the fat globule membranes; partly crystallised fat in neighbouring globules clumps, and the oil-in-water emulsion of cream inverts to the water-in-oil emulsion of butter (phase inversion). The liquid released is buttermilk. Cream is first aged cold so that the fat crystallises in the right proportion; churning temperature is adjusted by season, usually around 8 to 14 degrees C. The grains are washed, salted if desired, and worked to disperse the water into fine droplets. Ripened (cultured) cream butter is made from cream fermented with aroma-producing lactic cultures, whose diacetyl gives the typical flavour; sweet cream butter is made without ripening.
Under the Codex standard for butter, butter contains at least 80 per cent milk fat, not more than 16 per cent water and not more than 2 per cent milk solids-not-fat. Overrun in butter is the extra weight of butter over the weight of fat churned, expressed as per cent of fat: (butter minus fat) divided by fat, times 100. If 200 kg of 40 per cent cream (80 kg fat) gives butter of 81 per cent fat with no fat loss, butter = 80 / 0.81 = 98.8 kg and overrun = 23.5 per cent. In practice some fat is lost in buttermilk, and churning efficiency is judged by the fat in buttermilk.
Ghee is clarified milk fat made by heating butter or cream until the water boils off and the milk solids brown, then filtering. It is almost pure fat (around 99.5 per cent), with moisture of 0.5 per cent or less, and its characteristic flavour comes from compounds formed when protein and lactose in the curd are heated in fat: carbonyls, lactones, free fatty acids and Maillard products.
Four methods are taught. The indigenous (desi) method: milk is converted to dahi, churned to makkhan (country butter), and the makkhan is heated in an open pan. It gives good flavour but lower fat recovery and is hard to scale. The creamery butter method: unsalted butter is melted and heated to about 110 to 120 degrees C until frothing stops and the curd turns golden brown; the ghee is decanted and filtered from the residue. It is the commonest industrial method. The direct cream method: cream is heated directly to the clarification temperature; it saves butter making but takes more fuel and time and leaves more residue. The pre-stratification method: butter is melted at about 80 to 85 degrees C and left undisturbed for about 30 minutes so it separates into a top layer of floating curd particles, a middle fat layer and a bottom serum layer; the serum, which holds most of the water and SNF, is drained, and the remainder is heated to clarification temperature. Draining the serum saves fuel and gives ghee of lower acidity.
Ghee residue, the browned curd left behind, contains phospholipids that act as natural antioxidants. Ghee develops a granular texture when it cools slowly, because high-melting glycerides crystallise into grains; this graininess is prized in South Asian markets. Ghee spoils mainly by oxidation (tallowy flavour), accelerated by copper, iron, light and air, and by hydrolytic rancidity if moisture is left in it. Permitted antioxidants are listed in national regulations.
Ghee is a high-price fat and is adulterated with vegetable oils, hydrogenated vegetable fat (vanaspati) and animal body fats. Its purity is judged by analytical constants:
Reichert-Meissl (RM) value: mL of 0.1 N alkali needed to neutralise the steam-volatile, water-soluble fatty acids (mainly butyric and caproic) from 5 g of fat. Milk fat has a high RM value; vegetable fats and body fats have almost none, so adulteration lowers it.
Polenske value: the same for steam-volatile, water-insoluble fatty acids (mainly caprylic and capric). It is high for coconut oil, so a raised Polenske value with a lowered RM value points to coconut oil.
Kirschner value: a more specific measure of butyric acid.
Butyro-refractometer (BR) reading at 40 degrees C: rises with longer-chain and unsaturated fatty acids, so most vegetable oils raise it.
Iodine value (unsaturation) and saponification value (average chain length; milk fat is high because of its short-chain acids) complete the picture.
Normal values vary with species, region and feeding. Ghee from areas where cottonseed is fed heavily has a lower RM value and higher BR reading, and Indian standards have long carried separate limits for these cotton-tract areas. Specific tests add detail: the Baudouin test (furfural and hydrochloric acid give a pink-red colour with sesame oil) detects vanaspati in India, where sesame oil has been required in vanaspati as a marker; the Halphen test detects cottonseed oil; the phytosterol acetate test distinguishes vegetable fat (phytosterols) from milk fat (cholesterol); and opacity and crystallisation tests help with body fats. In India, ghee may also be graded under AGMARK, which is voluntary.
Ice cream is a frozen foam: air cells, fat globules and ice crystals dispersed in a concentrated unfrozen sugar solution. A typical mix contains about 10 to 16 per cent milk fat, 9 to 12 per cent MSNF, 12 to 16 per cent sugar and small amounts of stabiliser and emulsifier, but composition varies widely with style and law. The mix is blended, pasteurised (at higher conditions than milk), homogenised, cooled and aged at about 4 degrees C for some hours so that fat crystallises and the stabiliser hydrates. It is then frozen with whipping in a continuous freezer and drawn at about minus 5 to minus 6 degrees C, packed, and hardened quickly at very low temperature. Fast hardening keeps ice crystals small.
Overrun is the increase in volume due to incorporated air: (volume of ice cream minus volume of mix) divided by volume of mix, times 100. By weight, using the weight of a fixed volume: (weight of mix minus weight of the same volume of ice cream) divided by weight of the ice cream, times 100. A mix weighing 1.10 kg per litre made into ice cream weighing 0.55 kg per litre has 100 per cent overrun. Standards limit overrun indirectly by setting a minimum weight per litre and minimum total solids. Kulfi, the South Asian frozen dessert, is made from milk concentrated by boiling, sweetened and frozen in moulds without whipping, so it has little or no overrun and a dense texture.
Paneer and chhena are made by coagulating hot milk with acid, which precipitates casein together with much of the heat-denatured whey protein and traps the fat. Heating before acidification matters: denatured beta-lactoglobulin binds to casein and is carried into the curd, raising yield and giving the characteristic texture.
Paneer is a firm, sliceable, pressed product used in cooking. A widely taught method developed at the National Dairy Research Institute uses buffalo milk standardised to about 6 per cent fat and 9 per cent SNF, heated to about 90 degrees C, cooled to about 70 degrees C and coagulated with a 1 to 2 per cent solution of citric acid at about the same temperature, added slowly with stirring until the whey turns clear and greenish. The curd is held briefly, the whey drained, and the curd pressed in hoops, then immersed in chilled water to cool and firm it. Buffalo milk gives a firmer, more cohesive paneer and a higher yield than cow milk; cow milk paneer is softer and is often made with adjusted coagulation conditions.
Chhena is the softer, moister counterpart made mainly from cow milk and is not pressed hard. It is the base of the Bengali and Odia sweets rasogolla, sandesh and chhena poda. Cow milk is preferred because it gives a smooth, soft body that kneads well and holds its shape when cooked in syrup; buffalo milk chhena tends to be greasy and hard.
Yield depends on milk composition and recovery of solids. If 100 kg of milk yields 7 kg of total solids in the curd and the paneer is 55 per cent moisture, paneer weight is 7 / 0.45 = 15.6 kg; using the moisture instead of the solids fraction (7 / 0.55) is the common mistake.
Khoa (khoya, mawa) is milk partially dehydrated by continuous boiling with stirring and scraping in an open pan (karahi) until it becomes a semi-solid mass. It is the base of most North Indian milk sweets. Buffalo milk is preferred because it gives a higher yield, softer and smoother body, and better texture for sweets.
Three types are recognised by their use. Pindi khoa is compact and the driest, shaped into balls or pats, and used for burfi and peda. Dhap khoa is the softest and moistest and is used for gulab jamun and kalajamun. Danedar khoa has a granular texture produced by adding a small amount of acid (often citric acid or sour whey) to boiling milk, and is used for kalakand.
Khoa yield follows a solids balance. If 100 kg of milk containing 13 per cent total solids is concentrated to khoa of 70 per cent solids without loss, khoa = 13 / 0.70 = 18.6 kg and about 81.4 kg of water is evaporated.
Related concentrated products include rabri (layers of skin from slowly simmered sweetened milk), basundi (sweetened concentrated milk of western India), kheer or payasam (milk boiled with rice and sugar), and kalakand. Shelf life of khoa and its sweets is short at room temperature because of mould and yeast growth on the surface; hygienic handling, packaging and refrigeration extend it.
Fermented milks are made by the action of lactic acid bacteria (and sometimes yeasts) on milk. Lactic acid lowers the pH towards 4.6, the casein gels, and the bacteria add flavour compounds. Starters are grouped by temperature. Mesophilic starters (optimum about 25 to 30 degrees C) include Lactococcus lactis subsp. lactis and subsp. cremoris for acid, and the citrate-fermenting Lactococcus lactis subsp. lactis biovar diacetylactis and Leuconostoc species for flavour; their key aroma compound is diacetyl, the buttery note of dahi, cultured buttermilk and ripened-cream butter. Thermophilic starters (optimum about 40 to 45 degrees C) include Streptococcus thermophilus and Lactobacillus delbrueckii subsp. bulgaricus.
Dahi (curd) is the traditional South Asian fermented milk, made from boiled or pasteurised milk with a little of the previous day's dahi as inoculum, and incubated at ambient or warm temperature until set. Its flora is mixed and variable: mesophilic lactococci and leuconostocs with lactobacilli and sometimes S. thermophilus. Mishti doi of Bengal is sweetened, often caramel-coloured dahi. Chakka is dahi drained of whey in a cloth; shrikhand is chakka mixed with sugar and flavouring (cardamom, saffron), a western Indian product. Lassi is whipped dahi with water and sugar or salt; chhach is the diluted buttermilk left from churning dahi into makkhan.
Defects of dahi include wheying-off (syneresis) from over-acidification or disturbance after setting, gassiness from coliforms or yeasts, ropiness from some slime-forming strains, and bitterness from proteolysis. Bacteriophage attack on the starter is the main cause of slow or failed sets in industry; rotation of cultures and hygiene control it.
Under the Codex standard for fermented milks (CXS 243-2003), yoghurt is made with symbiotic cultures of Streptococcus thermophilus and Lactobacillus delbrueckii subsp. bulgaricus. The two grow better together than alone (protocooperation): the lactobacillus releases peptides and amino acids from casein that stimulate the streptococcus, and the streptococcus produces formic acid and carbon dioxide that stimulate the lactobacillus. Acetaldehyde is the key yoghurt flavour.
The mix (milk with added milk solids, sometimes sugar and stabiliser) is homogenised and heated hard, commonly 85 to 95 degrees C for several minutes, to denature whey proteins so that they bind to casein and give a firm gel with less syneresis. It is cooled to about 42 to 45 degrees C, inoculated at a few per cent, and incubated until pH falls to about 4.5 to 4.6, which takes a few hours, then cooled.
Other fermented milks: kefir (kefir grains containing lactic acid bacteria and yeasts; slightly alcoholic and fizzy), koumiss (traditionally mare milk; alcoholic), acidophilus milk (Lactobacillus acidophilus), Bifidobacterium-containing probiotic milks, ayran and doogh (salted diluted yoghurt drinks), labneh and skyr (strained), and the Nordic ropy milks.
Cheese is made by coagulating milk, separating the curd from the whey, and usually ripening the curd. Most cheeses are rennet-coagulated: chymosin, the main enzyme of calf rennet, cleaves kappa-casein at the Phe105-Met106 bond, releasing the hydrophilic caseinomacropeptide into the whey; the remaining para-kappa-casein micelles lose their stabilising layer and aggregate in the presence of calcium ions. Substitutes include fermentation-produced chymosin and microbial coagulants from fungi such as Rhizomucor miehei. Acid-coagulated fresh cheeses (cottage cheese, quark, cream cheese) are set by lactic acid alone, as are paneer-type cheeses by acid and heat.
The general sequence for a hard cheese such as Cheddar: standardise and pasteurise milk, add starter and rennet, cut the coagulum, cook (scald) the curd with stirring to expel whey, drain, cheddar (stack and turn the curd slabs as acid develops and the curd becomes fibrous), mill, salt, press and ripen. Pasta filata cheeses such as mozzarella are stretched in hot water once the curd reaches the right pH. Ripening is driven by residual coagulant, starter and non-starter bacteria and secondary flora: Penicillium roqueforti in blue cheeses, Penicillium camemberti on Camembert and Brie, Propionibacterium freudenreichii producing the eyes and sweet flavour of Swiss types.
Cheese yield is often estimated with the Van Slyke formula for Cheddar: yield per 100 kg milk = (0.93F + C minus 0.1) x 1.09 / (1 minus W), where F is fat per cent, C casein per cent and W the moisture fraction of the cheese. For milk of 3.6 per cent fat and 2.5 per cent casein making Cheddar with 37 per cent moisture, yield = (3.348 + 2.4) x 1.09 / 0.63 = about 9.9 kg. Late blowing of hard cheeses by Clostridium tyrobutyricum is prevented by bactofugation, nitrate (where permitted) or lysozyme.
A dairy package contains the product, protects it from microbes, light, oxygen, moisture and odours, carries legally required information, and makes distribution and use convenient.
Glass bottles are inert, transparent and returnable, but heavy and fragile, and clear glass lets light through. Low-density and linear low-density polyethylene films formed into pouches on form-fill-seal machines are the dominant package for pasteurised milk in India and much of Asia, because they are cheap and light; co-extruded films with a pigmented layer reduce light transmission. High-density polyethylene and PET bottles are common elsewhere for pasteurised and ESL milk, often pigmented. Aseptic cartons for UHT milk are laminates of paperboard (stiffness), polyethylene (sealing and liquid barrier) and a thin aluminium foil (light and oxygen barrier). Tinplate cans are used for evaporated and sweetened condensed milk, ghee and whole milk powder; lacquered tinplate prevents metal pick-up, which would catalyse fat oxidation. Flexible laminates with foil or metallised layers carry milk powder and ghee; parchment and foil laminates wrap butter; wax, plastic coatings and vacuum or shrink films protect cheese; coated paperboard cups and plastic tubs carry dahi, yoghurt and ice cream.
Milk exposed to sunlight or fluorescent light develops a burnt-feathers or cabbage-like off-flavour within hours. Riboflavin in milk acts as a photosensitiser; the excited riboflavin degrades the amino acid methionine to methional, the main light-induced flavour compound, and also promotes fat oxidation. Riboflavin itself is destroyed. Light-blocking pigments, foil layers and storage in the dark prevent it.
Oxygen promotes oxidised flavour in fat-rich products. Whole milk powder is packed under an inert gas (nitrogen, or nitrogen with carbon dioxide) to a low residual oxygen level, which greatly extends its shelf life. Ghee is filled hot into tins or jars with little headspace.
Aseptic packaging means filling a commercially sterile product into a sterile package in a sterile zone and sealing it. Packaging material is commonly sterilised with hot hydrogen peroxide followed by hot air to remove residues, sometimes with ultraviolet light; the filler zone is kept under sterile air overpressure.
Packaging must also meet food-contact and labelling rules (name, ingredients, net quantity, date marking, batch, manufacturer); in India these are set by FSSAI regulations, with Bureau of Indian Standards specifications for many packaging materials; other countries have equivalent rules.
Cleaning-in-place (CIP) cleans pipes, tanks, heat exchangers and fillers by circulating cleaning solutions without dismantling the equipment. The deposits are milk fat, protein and, on heated surfaces, milk stone, a hard scale of calcium phosphate and denatured protein.
A typical sequence is: pre-rinse with water to remove loose milk; circulate hot alkali (sodium hydroxide, commonly in the range of about 0.5 to 2 per cent, at around 70 to 80 degrees C) to dissolve protein and saponify fat; intermediate rinse; circulate acid (nitric or phosphoric acid, commonly below 1 per cent, at around 60 to 70 degrees C) to remove mineral scale, mainly on heated surfaces; final rinse; and sanitise just before use with hot water or a chemical sanitiser such as chlorine, iodophor or peracetic acid. Exact concentrations, temperatures and times are set by validation for each circuit.
Cleaning depends on four interacting factors, often called Sinner's circle or TACT: temperature, (mechanical) action, concentration and time. In pipes, mechanical action comes from turbulent flow; a velocity of at least about 1.5 metres per second is a common design rule.
Verification includes visual inspection, ATP bioluminescence swabs (which detect organic residues, not specific microbes), swab or rinse microbial counts, conductivity monitoring of solution strength and checks on final rinse water for residues. Biofilms in dead legs and worn gaskets are a common source of recurring contamination, so plant design avoids dead legs and uses hygienic fittings.
Hazard Analysis and Critical Control Point (HACCP) is the preventive food safety system described in the Codex General Principles of Food Hygiene (CXC 1-1969, revised 2020). It rests on prerequisite programmes (good hygiene practices: water, pest control, cleaning, personal hygiene, maintenance, supplier control) and applies seven principles: conduct a hazard analysis; determine critical control points (CCPs); establish critical limits; establish monitoring; establish corrective actions; establish verification; and establish documentation and record keeping. Codex sets these out within twelve steps, the first five being preparatory: assemble the team, describe the product, identify intended use and users, construct a flow diagram, and confirm it on site.
In a liquid milk plant the classic CCP is pasteurisation. The critical limit is the legal time-temperature combination (for example, at least 72 degrees C for 15 seconds); monitoring is the recording thermometer and flow control; the corrective action is automatic flow diversion and holding of suspect product; verification includes the alkaline phosphatase test, calibration of thermometers and review of charts. Other CCPs or control points include raw milk reception (antibiotic residues, temperature), metal detection, the pressure differential in regeneration, and in some products pH (fermented milks) or water activity. Hazards are biological (Listeria monocytogenes, Salmonella, pathogenic E. coli, Staphylococcus aureus enterotoxin, Bacillus cereus, Cronobacter in powdered infant formula), chemical (drug residues, aflatoxin M1, detergent residues, melamine) and physical (metal, glass).
Quality control tests products against specifications; quality assurance designs the system so that products meet them. A dairy laboratory runs raw milk reception tests, in-process checks (fat and SNF of standardised milk, pasteurisation records, CIP verification) and finished-product tests: composition, phosphatase for pasteurised products, coliforms and total counts, yeast and mould counts for butter, dahi and paneer, incubation tests for UHT, moisture and acidity for ghee, solubility and bulk density for powder, overrun and weight per litre for ice cream.
Sampling plans follow statistical principles: a two-class plan accepts a lot if no more than c of n sample units exceed the limit m; a three-class plan adds a higher limit M that no unit may exceed.
Chemical contaminants are monitored against limits. Aflatoxin M1 appears in milk when animals eat feed contaminated with aflatoxin B1; Codex sets a maximum level of 0.5 microgram per kg in milk, while the EU sets 0.05 microgram per kg, an example of how national limits differ. Veterinary drug residues are controlled against maximum residue limits set by Codex and national authorities, with withdrawal periods observed on the farm.
Shelf-life failure is investigated by asking where the organisms came from: spore formers that survive heat point to raw milk quality; Gram-negative psychrotrophs and coliforms point to post-pasteurisation contamination at the filler or in poorly cleaned pipes.
In India, food standards are set under the Food Safety and Standards Act, 2006, administered by the Food Safety and Standards Authority of India (FSSAI). It replaced the Prevention of Food Adulteration Act, 1954 and the Milk and Milk Products Order, 1992, among others. Product standards are in the Food Safety and Standards (Food Products Standards and Food Additives) Regulations, 2011, where dairy products form a separate section that was substantially revised in the late 2010s. The standards are amended often, so always confirm against the current consolidated text on the FSSAI website.
Milk is standardised by class. Minimum fat and SNF for buffalo, cow, goat or sheep and mixed milk differ by state or region in the schedule, reflecting real differences in local milk; buffalo milk minima are higher than cow milk minima. Processed milk classes are national and widely examined: toned milk at least 3.0 per cent fat and 8.5 per cent SNF; double toned milk at least 1.5 per cent fat and 9.0 per cent SNF; standardised milk at least 4.5 per cent fat and 8.5 per cent SNF; full cream milk at least 6.0 per cent fat and 9.0 per cent SNF; and skimmed milk not more than 0.5 per cent fat and at least 8.7 per cent SNF.
Other dairy products carry standards on moisture, fat (often expressed on a dry matter basis), acidity and microbiology. For example, paneer has a maximum moisture of 60 per cent and a minimum milk fat of 50 per cent of dry matter, with separate standards for low-fat paneer; khoa has a minimum milk fat on a dry matter basis; ghee has a maximum moisture of 0.5 per cent and limits for free fatty acids, BR reading and RM value that differ by state, with relaxed values for cotton-tract areas; ice cream has minimum total solids, milk fat, protein and weight per litre.
Voluntary marks sit alongside: Bureau of Indian Standards (BIS) standards for many dairy products, methods of test (for example the IS 1479 series for milk testing) and equipment; and AGMARK grading of ghee and butter under the Agricultural Produce (Grading and Marking) Act, 1937.
The Codex Alimentarius Commission, run jointly by FAO and WHO, writes international food standards. They are voluntary, but the World Trade Organization's Sanitary and Phytosanitary Agreement treats them as the reference for food safety, so countries that follow Codex are protected in trade disputes. Key dairy texts are the General Standard for the Use of Dairy Terms (CXS 206-1999), which defines milk, dairy products, composite milk products and reconstituted and recombined milk, and stops non-dairy products from using dairy names; individual standards for butter, milk fat products (including ghee under anhydrous milk fat and related products), milk powders, evaporated and sweetened condensed milks, cheeses, fermented milks (CXS 243-2003) and cream; the Code of Hygienic Practice for Milk and Milk Products (CXC 57-2004); maximum levels for contaminants such as aflatoxin M1 and melamine; and maximum residue limits for veterinary drugs.
National systems differ in emphasis. The European Union sets hygiene rules for raw milk (including limits for plate count and somatic cells) and for processing in Regulation (EC) No 853/2004, with official controls under its own regulation. The United States uses the Grade A Pasteurized Milk Ordinance, adopted by states, with the FDA's standards of identity for named products.
Learn the numbers that recur (63 degrees C for 30 minutes, 72 degrees C for 15 seconds, 0.009 g lactic acid per mL of 0.1 N alkali, Richmond's formula, freezing point near minus 0.53 degrees C, the FSSAI milk classes) and the one-reagent-one-colour adulterant tests. Practise a Pearson square, an added-water calculation and an overrun calculation until each takes under a minute. In the dairy practical, run the Gerber test, COB, alcohol test, MBRT and acidity yourself; you will be asked to interpret, not just recite, them. Visit a plant and trace milk from the dock to the pouch, naming each CCP.
A veterinarian five years into dairy work is usually the person a cooperative or chilling centre calls when milk is rejected. You decide whether a failed alcohol test means mastitis, colostrum or acidity, trace antibiotic-positive tankers to farms and withdrawal periods, advise on SCC and bacterial counts that cut the farmer's price, and check bulk milk coolers. In a plant you read pasteuriser charts, sign off phosphatase results, plan standardisation to legal classes, and investigate short shelf life by separating heat-resistant spore formers from post-pasteurisation contamination.
After twenty-five years the work is system design and regulation: setting raw milk payment schemes that reward hygiene, specifying UHT and powder lines, validating CIP and HACCP plans, approving packaging, and answering regulators and auditors. A senior dairy veterinarian in public service writes or enforces standards, plans surveillance for adulteration, aflatoxin M1 and residues, and decides when a product recall is needed. The judgement that matters most is which control is critical and which is ritual.
A researcher with long field experience works on questions the textbooks leave open: the heat stability and functionality of buffalo milk, process technology and shelf-life extension for paneer, khoa and dahi, characterisation and protection of traditional starter flora, rapid tests for synthetic milk and new adulterants, residue and contaminant risk assessment, and life-cycle and energy studies of chilling, UHT and drying. Such a researcher also advises Codex and national committees, where traditional South Asian products are still under-represented in international standards.
This manual gives exact figures only where they are fixed and widely taught (pasteurisation conditions, Gerber reagents, Codex butter composition, the FSSAI processed milk classes, paneer and ghee moisture limits, aflatoxin M1 levels). It deliberately gives ranges, not single values, for compositional averages, UHT and sterilisation conditions, CIP concentrations, khoa moisture by type, ice cream mix composition and the regional ghee constants, because these vary by source, country and plant. It does not reproduce the full FSSAI schedules (state-wise milk minima, ghee RM and BR limits, microbiological criteria, the many product standards) or the US PMO and EU microbiological limits, which change with amendments and should be read in the current official text. Not yet covered in depth: membrane processing beyond a mention (ultrafiltration, reverse osmosis), whey and casein products, infant formula, lactose-free milks, dairy effluent treatment, milk pricing schemes, and detailed instrumental methods (gas chromatography, infrared calibration, flow cytometry). Traditional products outside South Asia (for example African and Central Asian fermented milks) are named only briefly.
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.
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.