Food Science & Technology · Food Process Engineering · also Thermobacteriology, Canning Technology, Thermal Process Engineering, Heat Preservation of Foods
Designing and validating heat processes that make food safe and shelf-stable: microbial death kinetics, heat penetration, process calculation by the general and formula methods, retorts and aseptic systems, and the regulatory framework for low-acid canned foods. Agricultural Engineering's food-dairy-engineering introduces D, z and F; this subject teaches the full process-establishment practice.
In the exams: GATE XE-G Section 3 (thermal processing) and Section 4 (thermal sterilisation); ICAR AIEEA-PG Code 18 Unit II (thermobacteriology, sterilisation); ICAR AICE 35.a Unit 2 (process time evaluation for canned products)
Thermal processing is the use of heat to make a sealed food safe and stable. A can of beans, a jar of tomato sauce, a retort pouch of curry and a carton of UHT soup all rest on the same idea: heat the food, in or before its container, long enough at a high enough temperature to destroy the microorganisms that could harm the consumer or spoil the food, and then keep new ones out. Nicolas Appert showed in the early nineteenth century that food heated in sealed bottles kept for months; Louis Pasteur later explained why; and in the 1920s Bigelow, Ball and their colleagues turned the practice into a quantitative science with the D value, the z value and the lethality integral. That science is what this subject teaches.
The subject matters because the margin for error is unusually small and the hazard unusually severe. Clostridium botulinum, a spore-forming anaerobe, can grow in a low-acid food sealed away from air and produce one of the most potent toxins known. A process that falls short by a few minutes, a can with a faulty seam, or a retort that was not properly vented can cause illness and death. At the same time, every extra minute of heat damages colour, texture, flavour and vitamins. The food technologist's job is therefore twofold: establish a process that is demonstrably safe, and make it no more severe than it needs to be. Doing both requires kinetics (how fast heat kills organisms and destroys quality), heat transfer (how fast the slowest-heating point of the container gets hot), container science (how the package keeps its seal), equipment knowledge (how retorts and aseptic systems deliver heat) and regulation (who may set a process, and what records prove it was delivered).
Worldwide, the topic is examined in two ways. Entrance and postgraduate exams such as GATE XE-G and XL-U (Food Technology), ICAR AIEEA-PG Code 18 and ICAR AICE-JRF/SRF 35.a test the definitions and, above all, the calculations: D and z values, lethal rates, F0, log reductions, spoilage probability, and process time by the general and formula methods. In professional life the same ideas appear as heat-penetration studies, process filings, retort records, seam teardown reports and deviation decisions. This manual covers both. It builds on the short survey of D, z and F in the Agricultural Engineering subject food-dairy-engineering and goes further into process establishment, containers, retorts, aseptic systems and regulation. Values quoted for organisms are typical textbook values; real processes must use data validated for the specific product, container and equipment.
When a suspension of bacterial spores is held at a constant lethal temperature, the number of survivors falls in a pattern that, for design purposes, is treated as first order: in each equal interval of time the same fraction of the remaining population is destroyed. Plotting the logarithm of the survivors against time therefore gives a straight line, called the survivor curve. The time needed for this line to fall by one log cycle, that is for 90 per cent of the population to be destroyed, is the decimal reduction time, D. It is always quoted with its temperature, as D at 121.1 °C, D121.1, or D at 250 °F.
The survivor equation is log(N0/N) = t/D, where N0 is the initial number, N the number after time t. Rearranged, t = D × log(N0/N): the time needed is D multiplied by the number of log reductions wanted. If a population of 100,000 spores must be reduced to 1, five log reductions are needed, and at a temperature where D is 2 minutes this takes 10 minutes. The first-order rate constant k is related to D by k = 2.303/D, because 2.303 is the natural logarithm of 10.
D depends on three things: the organism (and its strain and history), the medium in which it is heated, and the temperature. Low pH, salt and some preservatives tend to lower heat resistance; fat, low water activity and high concentrations of sugar or protein tend to protect spores and raise D. A D value is therefore only meaningful for a stated organism, medium and temperature. Real survivor curves are not always straight: they can show an initial shoulder, when clumped spores or spores that need heat activation delay the decline, and a tail, when a small, more resistant fraction survives longer than the line predicts. Process design uses the straight-line model with safety margins and validation in the actual product.
D falls sharply as temperature rises. If log D is plotted against temperature, the result over the range used in food processing is close to a straight line, the thermal death time (TDT) or phantom curve. The temperature rise that changes D by a factor of ten — moves it by one log cycle — is the z value. So if D is 4 minutes at 111 °C and z is 10 °C, D is 0.4 minutes at 121 °C and 0.04 minutes at 131 °C.
Mathematically, log(D1/D2) = (T2 − T1)/z, and the D at any temperature T follows from a reference value: D_T = D_ref × 10^((T_ref − T)/z). The same relation applies to thermal death times. A thermal death time F_T is the time at temperature T needed to achieve a stated destruction, and F_T = F_ref × 10^((T_ref − T)/z).
Typical z values differ greatly between kinds of reaction, and this difference is the foundation of modern process design. Bacterial spores usually have z values of about 7 to 12 °C, and 10 °C is the conventional value for Clostridium botulinum in low-acid foods. Vegetative cells often have z values of 4 to 8 °C. Quality attributes such as colour, texture and vitamins usually have much higher z values, commonly 20 to 50 °C, and many enzymes fall in between. Because spores are far more sensitive to a temperature rise than quality is, raising the process temperature and shortening the time gives the same microbial kill with less quality loss. This is the principle behind high-temperature short-time and ultra-high-temperature processing.
Chemists describe the temperature dependence of reaction rates with the Arrhenius equation, k = A exp(−Ea/RT), where Ea is the activation energy, R the gas constant and T the absolute temperature. Food scientists use the z value because it is simpler to apply in processing calculations. Over the narrow range of temperatures used in a given process, the two models agree closely, and they can be converted: Ea ≈ 2.303 R T1 T2 / z, with T1 and T2 in kelvin at the ends of the range considered. Because T1 T2 changes with temperature, a single z corresponds strictly to a single activation energy only over a limited range, which is why z values should not be extrapolated far beyond the data they came from.
A third description, used mostly for storage and quality reactions, is Q10, the factor by which a rate increases for a 10 °C rise. Q10 = 10^(10/z), so a z of 10 °C means a Q10 of 10, while a quality reaction with a z of 33 °C has a Q10 of only about 2. In exam questions it is worth checking which description a problem uses and converting carefully, keeping track of the sign: a higher temperature always means a smaller D and a shorter thermal death time.
A real container does not sit at one temperature: its contents heat up, hold and cool down. To add up the killing effect of a changing temperature, every minute at temperature T is converted into equivalent minutes at a reference temperature. The conversion factor is the lethal rate, L = 10^((T − T_ref)/z). One minute at the reference temperature has L = 1; one minute at 10 °C below it, with z = 10 °C, has L = 0.1; one minute at 10 °C above it has L = 10.
The integrated lethality of a process is its F value: F = ∫ L dt, the sum of lethal rates over the whole time-temperature history. It is expressed in minutes at the reference temperature and with the z value used, written F with subscript temperature and superscript z. For low-acid canned foods the convention is a reference of 121.1 °C (250 °F) and z = 10 °C (18 °F), and the F value on this basis is written F0. An F0 of 3 minutes means the process is as lethal to an organism with z = 10 °C as 3 minutes at 121.1 °C followed by instant cooling would be.
The F value needed for a given log reduction is F = D_ref × number of log reductions, provided D and F are on the same reference temperature. This links the kinetics of chapter 1 to the process: if the target organism has D121.1 = 0.5 min and a 6-log reduction is required, the process must deliver F0 = 3 min at the slowest-heating point.
For low-acid foods the public-health target is Clostridium botulinum. The classical criterion, the 12D concept, is that the process should reduce the population of the most heat-resistant C. botulinum spores by twelve log cycles. The reasoning is probabilistic: even if a container held one spore, the chance that it survives a 12D process is one in 10^12. Using the textbook value D121.1 ≈ 0.21 min for proteolytic C. botulinum spores, 12D corresponds to F0 ≈ 2.5 min, and a minimum F0 of about 3 minutes is widely quoted as the botulinum cook.
In practice most commercial processes for low-acid foods deliver more than this, often several times more, because spoilage organisms are more heat resistant than C. botulinum. Mesophilic putrefactive anaerobes such as Clostridium sporogenes (often studied through the strain known as PA 3679) and thermophiles such as Geobacillus stearothermophilus have higher D values. Processes are therefore commonly designed to give the botulinum safety target and a separate spoilage target, for example a stated number of log reductions of a mesophilic spoilage organism, with the thermophile risk managed by keeping products below thermophile growth temperatures during storage or by giving still more heat for products sold in hot climates. The exact F0 values used vary with product, container, initial contamination and market, and they are set by a process authority for each product, not taken from a general table.
Because an F value depends on the reference temperature and z, two F values can only be compared when both are stated on the same basis. F0 is the standard for sterilisation of low-acid foods. For pasteurisation of acid foods, milk or juices, lower reference temperatures are used (for example 65 °C, 70 °C, 85 °C or 93.3 °C, depending on product and organism) and different z values; these are written with their reference and z explicitly.
A useful check in any calculation is that a process held at a temperature below the reference must deliver an F smaller than the time held, and one above the reference must deliver more. A constant hold of 30 minutes at 111.1 °C with z = 10 °C has a lethal rate of 0.1 and delivers only F0 = 3 min, while 30 seconds at 131.1 °C has a lethal rate of 10 and delivers F0 = 5 min. Students who reverse the sign of the exponent produce answers in which a cooler process is more lethal — a sign that the calculation has gone wrong.
The severity of the process a food needs depends first on its acidity. Clostridium botulinum does not grow or produce toxin at pH 4.6 or below, under the conditions found in properly made foods. Foods with an equilibrium pH above 4.6 can therefore support the organism if spores survive, and must be sterilised to destroy them; foods at or below pH 4.6 are protected by their acidity, and need only enough heat to destroy the less resistant organisms that can grow at low pH — vegetative bacteria, yeasts, moulds and some acid-tolerant spore-formers.
In United States regulation, which many other countries follow closely, a low-acid food is one with a finished equilibrium pH greater than 4.6 and a water activity greater than 0.85, excluding alcoholic beverages; an acidified food is a low-acid food to which acid or acid foods are added to bring the finished equilibrium pH to 4.6 or below. Water activity matters because C. botulinum cannot grow at low water activity either, so a food that is dry enough is not treated as low-acid even if its pH is high. Equilibrium pH means the pH after acid has had time to penetrate the particles, not just the pH of the brine on the day of packing.
Classification schemes in textbooks vary. A common one divides canned foods into low-acid (pH above 4.6: most vegetables, meat, fish, milk, soups), acid (about pH 3.7 to 4.6: many fruits, tomatoes, pears) and high-acid (below about 3.7: citrus, berries, pickles, sauerkraut). Some older texts add a medium-acid band between 4.6 and about 5.0. The legal line that matters for botulism control is 4.6.
In low-acid foods the public-health target is C. botulinum, especially its heat-resistant proteolytic types, and the spoilage targets are mesophilic and thermophilic spore-formers. Mesophilic putrefactive anaerobes such as C. sporogenes cause swollen, putrid cans. Thermophilic flat-sour organisms such as Geobacillus stearothermophilus produce acid without gas, so the can looks normal but the food is sour. Thermophilic anaerobes not producing hydrogen sulphide, such as Thermoanaerobacterium thermosaccharolyticum, produce hydrogen and carbon dioxide and swell cans; sulphide spoilage organisms (Desulfotomaculum nigrificans) blacken the food and give a rotten-egg smell without swelling. Thermophile spoilage usually happens when cans are cooled too slowly or stored hot.
In acid foods the targets are organisms that can grow at low pH: Bacillus coagulans (flat sour in tomato products), butyric anaerobes such as Clostridium pasteurianum, heat-resistant moulds such as Byssochlamys, and, in fruit juices, Alicyclobacillus species. These require far milder processes than low-acid foods, often temperatures near 90 to 100 °C rather than 115 to 125 °C. In high-acid foods only yeasts, moulds and lactic acid bacteria normally matter, and hot filling or short pasteurisation is enough.
Adding acid can move a food from the low-acid class into the acidified class and so allow a pasteurisation instead of a full sterilisation process. Pickled vegetables, some sauces, peppers and artichoke hearts are examples. The control only works if the equilibrium pH reaches 4.6 or below throughout the product, quickly enough that C. botulinum spores cannot germinate in the meantime. Large or dense pieces acidify slowly; oil layers can shield particles; and a few products buffer strongly. For this reason acidified foods are themselves regulated, with requirements for scheduled processes, pH measurement and records, as described in chapter 12.
In practice, acidification is checked by measuring the pH of blended product, not just of the covering liquid, at intervals after packing until it is stable, and by controlling the ratio of acid brine to solids at the filler. Where products are near the boundary, processors commonly aim well below 4.6 to allow for measurement error and natural variation in raw materials. Acidification also changes flavour, so it is used only where the product can carry the sour taste or where the acid is part of its traditional character.
A process is only as good as the heating of the slowest-heating point in the container, called the cold spot or slowest heating zone. Where that point lies depends on how heat moves through the food. In conduction-heating foods — solid packs such as corned beef, pumpkin purée, thick pastes and many pet foods — the food does not move and heat travels inward by conduction; the cold spot is the geometric centre of the container, and heating is slow. In convection-heating foods — thin liquids such as broths, juices and vegetables in thin brine — warmed liquid rises along the walls and cooler liquid sinks in the centre, so heating is faster and the cold spot lies on the vertical axis below the geometric centre, in the lower part of the can. Many foods heat by a mixture of the two, and some change mode during the process: a starchy product may heat by convection until starch gelatinises and then by conduction, giving a broken heating curve with two straight-line segments.
Container size and shape matter. Larger containers heat more slowly; flat containers such as pouches and trays heat faster than cylinders of the same volume because the path from the surface to the centre is shorter. Agitation during processing, by rotating the container, greatly speeds heating of liquid and particulate foods by mixing the contents and moving the headspace bubble.
Two kinds of test are used to establish a process. A temperature distribution test maps the heating medium in the retort: temperature sensors are placed throughout a loaded retort to find the slowest-heating location in the equipment and to show that venting and come-up procedures give uniform temperature. A heat penetration test measures the temperature inside the food: thermocouples or wireless data loggers are fitted so that the sensing tip is at the container's cold spot, the containers are filled and processed exactly as in production, and the temperature is recorded throughout heating and cooling.
Heat penetration tests are done with the worst-case conditions expected in production: the lowest initial temperature, the maximum fill weight, the thickest consistency, the largest particle size and the slowest-heating position in the retort. Several containers are measured in several runs, and the slowest-heating results are used for design. The cold spot itself is first located by placing sensors at several heights on the axis in trial containers. Errors in fitting the sensor — heat conducted along the thermocouple wire, the tip not at the cold spot, or the fitting changing how the food heats — are a known source of error and are minimised by using fine wires, proper receptacles and checks.
Heat penetration data are analysed by plotting the logarithm of (TR − T), where TR is the retort temperature and T the cold-spot temperature, against time. After an initial lag, the plot becomes a straight line. Its slope gives fh, the heating rate index: the time in minutes for the straight portion to pass through one log cycle, that is for the temperature difference to fall tenfold. A large fh means slow heating.
The straight line, extended back to the start of the process, meets the time axis at a pseudo-initial temperature T_pih, which is lower than the true initial temperature because of the lag. The heating lag factor is jh = (TR − T_pih)/(TR − T_ih), where T_ih is the actual initial temperature. A value of jh near 1 is typical of convection heating; conduction heating in cylinders gives values approaching 2 at the centre. The cooling curve is analysed in the same way to give fc and jc. In Ball's method the process time is measured from the moment the retort reaches its processing temperature, and about 42 per cent of the come-up time is taken as effective at retort temperature, so the zero of the corrected time scale is set accordingly. These parameters are the inputs to the formula method of chapter 6.
The general method, introduced by Bigelow and colleagues in 1920, calculates the lethality of a process directly from the measured cold-spot temperatures. Each temperature reading is converted to a lethal rate, L = 10^((T − 121.1)/z), and the lethal rates are integrated over time. The result is the F0 delivered by that process. Because it uses the actual time-temperature history, the method needs no assumptions about the shape of the heating curve, and it automatically includes the lethality contributed during come-up and cooling.
In its simplest graphical form, lethal rate is plotted against time and the area under the curve is measured; the area is F. Today the integration is done numerically. With readings at equal intervals Δt, the trapezoidal rule gives F ≈ Δt × (L1/2 + L2 + L3 + … + L_(n−1) + L_n/2), and Simpson's rule can be used for an odd number of points. Short reading intervals — commonly a minute or less — keep the error small, especially near the peak where L changes fastest.
Suppose the cold spot of a can reads 105, 112, 117, 120, 121, 121, 119 and 110 °C at one-minute intervals. With z = 10 °C the lethal rates are 0.025, 0.123, 0.389, 0.776, 0.977, 0.977, 0.617 and 0.078. The trapezoidal sum is 1 × (0.025/2 + 0.123 + 0.389 + 0.776 + 0.977 + 0.977 + 0.617 + 0.078/2) ≈ 3.91 min. A student who credits only the two minutes near 121 °C would find about 2 min and underestimate the process badly, because much of the lethality comes from the minutes just below the peak.
Typical errors in exams are: reversing the sign of the exponent; using a natural logarithm or e instead of 10; using the retort temperature rather than the cold-spot temperature; forgetting the half-weight of the end points in the trapezoidal rule; and mixing seconds and minutes. Another important point is that readings below about 100 °C contribute almost nothing to F0 with z = 10 °C (L at 100 °C is about 0.008), so including or excluding them changes the answer very little.
The general method gives the lethality of a process that has already been run. To find the process time needed for a target F0, the heating part of the curve is shifted along the time axis — in effect assuming that heating continues longer at retort temperature before cooling begins — until the integrated area equals the target. The improved general method of Patashnik and later computer methods do this numerically. Modern practice usually combines the general method, applied to data from heat penetration tests, with a formula or numerical model to predict the effect of changes in retort temperature or initial temperature. Because the method depends entirely on the measured data, its reliability depends on the quality of the heat penetration test: the slowest container, measured at its true cold spot under worst-case conditions.
Ball's formula method, published in 1923 and later refined, predicts process time from the heating parameters fh and jh and a table relating the process to its lethality. Its advantage over the general method is that once fh and jh are known for a product and container, the process time for any retort temperature, initial temperature and required F can be calculated without a new heat penetration test, provided conditions stay within the range where the parameters are valid.
The heating equation for the straight part of the heating curve is log(TR − T) = log[jh(TR − T_ih)] − t/fh. At the end of heating, the cold point is g degrees below the retort temperature (g = TR − T at the end of heating). Rearranging gives Ball's process time: B = fh × log(jh × Ih / g), where Ih = TR − T_ih is the initial temperature difference. B is measured from the corrected zero, which credits about 42 per cent of the retort come-up time as effective. The operator's process time at retort temperature, Pt, is B minus 0.42 times the come-up time.
The value of g that gives the required lethality is found from Ball's tables. The required F at reference temperature is first converted to U, the equivalent minutes at the retort temperature: U = F × 10^((121.1 − TR)/z). For a retort temperature below 121.1 °C, U is larger than F. The ratio fh/U is then calculated, and the table, prepared for the appropriate z value and cooling lag factor, gives the corresponding g (often as log g). Finally B = fh × (log(jh × Ih) − log g).
Example: a product has fh = 40 min and jh = 1.4, initial temperature 70 °C, retort at 121.1 °C, and the required F0 is 6 min. Then U = 6 min, fh/U = 6.67; suppose the table gives log g = 0.70 for this fh/U at z = 10 °C. Then Ih = 51.1 °C, log(jh Ih) = log(71.54) = 1.855, and B = 40 × (1.855 − 0.70) = 46.2 min. With a 10-minute come-up, Pt = 46.2 − 4.2 = 42.0 min. In exams the table value is normally given in the question, and the skill tested is the chain of conversions.
Ball's original method assumes the cooling lag factor is about 1.41 and that the heating curve is a single straight line. Errors arise when cooling lag differs, when the curve is broken, or when the product heats by conduction and the cooling contributes much of the lethality. Stumbo's method uses more extensive tables covering different cooling lag factors and z values, and broken-heating formulas exist for curves with two slopes. Numerical models solving the heat conduction equation, and computer programs that apply the general method to simulated temperatures, are now common in industry. Whatever the method, a calculated process must be confirmed against heat penetration data, and the process authority chooses the method and its safety margins. The formula method is still taught and examined because it shows clearly how heating rate, lag, initial temperature and retort temperature each affect process time: a higher initial temperature or retort temperature shortens B; a larger fh lengthens it in direct proportion.
A canned low-acid food is not absolutely sterile. It is commercially sterile: heat has made it free of microorganisms capable of growing in the food under the normal non-refrigerated conditions of storage and distribution, and free of viable microorganisms of public-health significance. Spores of thermophiles may survive a commercial process, because destroying them would require so much heat that the food would be ruined; they stay dormant as long as the product is not held hot. That is why rapid cooling after processing, and storage below thermophile growth temperatures, form part of the safety and quality system.
Commercial sterility is checked in practice by incubating samples and examining them. A common scheme holds containers at about 30 to 37 °C to reveal mesophilic survivors and at about 55 °C to reveal thermophiles, for periods commonly in the range of one to two weeks for mesophiles and several days for thermophiles, followed by inspection for swelling, pH change and microscopic examination. Exact conditions are set by national methods and should be taken from them.
The survivor equation turns a process into a probability of spoilage. If each container starts with N0 spores of an organism whose D at the reference temperature is D_ref, a process of lethality F leaves N = N0 × 10^(−F/D_ref) survivors per container. When N is much less than one, it can be read as the probability that a given container still holds a viable spore, that is the expected fraction of containers that spoil. For example, if each can holds 100 spores of an organism with D121.1 = 1.2 min and the process delivers F0 = 6 min, then N = 100 × 10^(−5) = 10^(−3), and about one can in a thousand would be expected to spoil from that organism.
The same reasoning shows what drives spoilage risk: the initial load, the heat resistance and the delivered lethality. Halving the initial load halves the expected number of spoiled cans; adding one D of lethality reduces it tenfold. This is why good raw-material hygiene, short holding times between filling and retorting, and control of blanchers and fillers, where thermophiles can multiply, matter as much as the retort itself. Processes are commonly designed so that the predicted spoilage rate from the target spoilage organism is very small, often of the order of one container in tens of thousands to one in a million or less; the acceptable value is a business and regulatory decision for each product.
When cans spoil, the pattern tells the investigator what went wrong. Swollen containers are graded by severity: a flipper looks normal but one end flips out when the can is struck; a springer has one end bulged that can be pushed back, making the other bulge; a soft swell has both ends bulged but they yield to pressure; a hard swell cannot be pushed in at all. Swelling can be caused by microbial gas, by hydrogen produced when acid foods corrode the tinplate (hydrogen swell), or by overfilling.
Microbiology then separates the main causes. Underprocessing usually leaves a pure culture of heat-resistant spore-formers, often of one type. Leakage after processing — through faulty seams or contaminated cooling water drawn in while the can contracts — usually gives a mixed flora including non-spore-formers such as cocci and short rods that could never survive the process. Thermophilic spoilage such as flat sour points to slow cooling, hot storage, or build-up of thermophiles in the plant before retorting. Spoilage with no viable organisms but evidence of growth suggests pre-process spoilage: microbes grew and died before the food was retorted. Correct diagnosis is essential because the remedies are completely different: a longer process does not cure leaker spoilage, and better seams do not cure underprocessing.
Most canned foods use metal containers. Three-piece cans have a body rolled from a flat sheet and welded along a side seam, with two ends attached by double seams. Two-piece cans have a body drawn from a single disc (drawn-and-redrawn or drawn-and-wall-ironed) with one end seamed on, so there is no side seam. The metal is usually tinplate (steel coated with tin), tin-free steel (steel with chromium and chromium oxide layers), or aluminium. The inside is usually lacquered or enamelled to prevent the food reacting with the metal, to avoid discoloration such as black iron or tin sulphide stains in sulphur-rich foods, and to limit metal pickup. Can ends are often beaded and have expansion rings so that they can flex with internal pressure changes during processing.
Glass jars and bottles are inert and transparent but heavier and more fragile, and glass must be heated and cooled more gently to avoid thermal shock; jars are usually processed with overriding air pressure to keep closures in place. Closures are lug or press-on twist caps with a sealing compound, often with a safety button that shows when the vacuum is lost.
A double seam joins the can end to the body by interlocking five layers of metal — three of the end and two of the body — with a sealing compound filling the spaces. It is formed in two operations on a seamer. The first-operation roll curls the edge of the end (the cover hook) under and around the flange of the body (the body hook); the second-operation roll flattens and tightens the interlocked hooks so that they are pressed together with the compound sealed between them. A seaming chuck inside the end supports it against the rolls and a base plate holds the can.
The quality of a double seam is described by external measurements — seam thickness, seam width (also called seam length or height), and countersink depth — and by internal ones, seen when the seam is torn down or sectioned: body hook length, cover hook length, overlap, tightness (judged by wrinkles on the cover hook) and the condition of the juncture at the side seam. Overlap, the length over which body hook and cover hook overlap, is the key internal measure. A widely used approximation is: overlap = cover hook + body hook + 1.1 × end-plate thickness − seam width. Body hook butting, the proportion of the internal seam occupied by the body hook, is also calculated. Acceptable ranges are set by the can maker for each can size and must be met.
Seams are inspected visually at short intervals during production, commonly at least every half-hour on each seaming head, and torn down or sectioned at longer intervals, with all measurements recorded. Visual faults include false seams (hooks not interlocked), droops at the side seam, vees, cut-overs (metal fractured over the chuck), sharp seams, spurs, and deadheads (the chuck slipped and the seam was not formed). Teardown measures hooks and overlap and grades wrinkles to judge tightness. Seam faults are dangerous because a leak lets cooling water or air carry microorganisms into the sterilised food, and leakage is a classic cause of spoilage and of some botulism outbreaks.
Retort pouches and semi-rigid trays and bowls are made of multilayer laminates. A typical retortable pouch combines an outer polyester layer for strength and printing, an aluminium foil or a high-barrier polymer layer for oxygen and light barrier, and an inner polypropylene sealant layer that withstands retort temperatures. Their thin profile gives faster heat penetration than cans of the same volume, shortening the process and improving quality. They also bring their own risks: seal integrity must be checked (seal strength, burst tests, visual seal inspection for channels, wrinkles and contamination in the seal), the pouches must be held in racks that keep their thickness uniform, and the retort must provide overriding pressure, because the pouch cannot resist the internal pressure that develops when its contents are hotter than the boiling point at atmospheric pressure.
The simplest retort is a still batch pressure vessel, vertical or horizontal, into which crates or baskets of containers are loaded and then heated with saturated steam. The most important operating step in a steam retort is venting: at the start of each cycle, air must be driven out through vents, because a steam-air mixture at the same pressure is cooler than pure steam, and pockets of air create cold zones where cans are underprocessed. Venting schedules — the time and temperature the vents must stay open — are established by temperature distribution tests for each retort and loading pattern, and they are part of the scheduled process.
After the hold, containers are cooled, usually with water introduced under air pressure. Pressure cooling prevents the internal pressure of hot cans from straining the seams when the steam pressure falls, and chlorinated or otherwise treated cooling water is used because cans can draw in water through microscopic seam gaps while they contract during cooling. Retorts are fitted with an accurate reference thermometer or temperature sensor, a temperature recorder, a pressure gauge, steam controllers, vents, bleeders that let steam flow continuously and carry away air and condensate, and safety valves.
Water-immersion and water-spray (or cascading water) retorts, and steam-air retorts with fans, heat containers using water or a steam-air mixture under an overriding air pressure that is independent of temperature. They are used for glass, pouches, trays and other containers that cannot withstand the pressure differences of a pure steam process.
Rotating the containers during processing mixes the contents and so speeds heating for liquid and particulate foods. In end-over-end rotation the container turns about an axis perpendicular to its own, and the headspace bubble travels through the product, giving strong mixing; in axial rotation the container rolls about its own axis. Batch rotary retorts rotate the whole load in a cage; continuous rotary cookers (reel-and-spiral types) carry cans through a cylindrical shell on a rotating reel while a spiral moves them along, so that each can rolls on the shell wall. Agitation is effective only when the product can move: fill, headspace, consistency and rotation speed are critical factors that must be controlled because a smaller headspace or a thicker product slows heating. For conduction-heating solid packs, rotation gives little benefit.
A hydrostatic steriliser is a tall continuous system in which containers are carried by a chain through a water leg, a steam dome and a second water leg. The steam pressure in the dome is balanced by the height of water in the legs, so no mechanical pressure lock is needed: the containers enter through hot water, are preheated as they descend, processed in steam at the temperature set by the steam pressure, and cooled as they rise through the outlet leg. The water column needed equals the gauge pressure of the steam: for steam at about 121 °C, the gauge pressure is about 1.05 bar, which needs a water column of roughly 10.7 m, so the towers are very tall. Hydrostatic sterilisers suit very large outputs of a single container size and give gentle handling with low steam and water use.
Other continuous systems include continuous rotary cookers mentioned above, and newer designs using pressure locks or flame sterilisation. Flame sterilisation, in which cans are rotated directly over gas flames at atmospheric pressure, gives very rapid heating of convection-heating products in small, strong cans; it relies on internal pressure being resisted by the can itself.
In aseptic processing the food and the package are sterilised separately and brought together in a sterile environment. The food, usually a pumpable liquid or a liquid with particles, is heated rapidly in a heat exchanger — plate, tubular or scraped-surface, or by direct steam injection or infusion — held at a high temperature in a holding tube for a short, defined time, and then cooled before filling. Because the product is heated as a thin layer or in turbulent flow, temperatures of about 135 to 150 °C can be reached in seconds, and the high-temperature short-time combination gives the required lethality with much less quality damage than in-container sterilisation. Packages such as cartons, bags, cups and bottles are sterilised, commonly with hydrogen peroxide followed by heat to remove the residue, or with heat, irradiation or other agents, and filled and sealed in a zone kept sterile by filtered overpressure air.
The process is designed on the holding tube. Lethality is credited only for the time spent in the holding tube at the measured outlet temperature, not during heating and cooling, which adds a safety margin. The critical time is the residence time of the fastest-moving particle or fluid element, not the average. In laminar flow of a Newtonian fluid in a pipe, the maximum velocity is twice the mean velocity; in turbulent flow it is commonly taken as about 1.2 times the mean. The holding tube length is therefore at least the maximum velocity multiplied by the required hold time, and the tube must slope upward so that it stays full and air cannot shorten the residence time.
When the product contains particles, the centre of the particle heats more slowly than the surrounding liquid, and the fastest particle may pass through the holding tube in less than the mean time. Process establishment for particulate aseptic products must account for both: the slowest-heating particle travelling at the shortest residence time. This needs data on particle residence time distribution and fluid-to-particle heat transfer, and is one of the most demanding validations in food processing. Many regulators require that such processes be filed with detailed supporting evidence.
Before production, the whole downstream system — pipework, valves, surge tanks and filler — is sterilised, typically with hot water or steam at a defined temperature and time, and kept sterile with steam barriers on valves and seals. Loss of temperature in the holding tube must divert flow automatically through a flow-diversion valve so that underprocessed product never reaches the filler. Records include holding-tube outlet temperature, flow rate, pressure differentials in regenerators, and package-sterilant concentration and temperature.
Foods with an equilibrium pH of 4.6 or below need milder processes because C. botulinum cannot grow in them. The targets are the most resistant organisms able to grow at the food's pH: vegetative bacteria, yeasts, moulds, heat-resistant mould spores and acid-tolerant spore-formers. Processes use lower reference temperatures, commonly between about 65 °C and 100 °C depending on the product and organism, and are expressed as pasteurisation values with their reference temperature and z stated.
Common methods are hot filling, in which the product is heated, filled while hot and the container inverted or held so that the hot product treats the closure and headspace before cooling; in-container pasteurisation in tunnels or water baths; and HTST treatment with aseptic or clean filling. Fruit juices in some jurisdictions must achieve a stated log reduction of the most resistant pathogen of concern — for example, United States juice HACCP rules require a 5-log reduction. Acidified foods must also reach their target equilibrium pH quickly; their scheduled process sets maximum particle size, acid-to-product ratio, pH limits and the heat treatment. Tomato products, whose pH can lie near 4.2 to 4.6, sometimes need acidification and closer control because Bacillus coagulans flat sour and, at the upper end, the botulinum boundary are close.
Heat that destroys spores also cooks the food: it softens texture, changes colour and flavour, and destroys heat-sensitive nutrients such as thiamin and vitamin C. To compare processes for quality, the same integration used for lethality is applied with quality kinetics. The cook value, C, is the integrated lethal rate calculated with a z value typical of quality changes and a reference temperature of 100 °C; a z of 33 °C is a common convention, giving C with reference 100 °C and z 33 °C. A process with a higher C cooks the food more.
Because quality reactions have high z values, their rate increases only modestly with temperature. Raising temperature by 10 °C increases the lethal rate of spores (z = 10 °C) tenfold but the rate of a quality reaction with z = 33 °C only about twofold. To give the same F0 in a tenth of the time, the quality damage falls to roughly a fifth. This is the quantitative reason for HTST and UHT processing, for agitating retorts and for thin-profile packages, all of which allow higher temperatures or faster heating.
In conduction-heating containers the picture is more complicated. The process must deliver the target F0 at the centre, but the food near the wall is overcooked because it spends the whole process close to retort temperature. Raising the retort temperature shortens the process but makes the surface layer hotter; lowering it reduces surface damage but lengthens the process. For a given container and food there is therefore an optimum retort temperature that maximises average nutrient retention, and calculations commonly place it in the range of about 115 to 125 °C for many conduction-heated foods, depending on the container size and the quality factor. For convection-heating and agitated products, and for aseptic processing, where the whole product reaches nearly the same temperature, higher temperatures almost always improve quality.
Ways to improve quality while keeping safety include: reducing container thickness (pouches, trays), agitation, higher temperatures for convection products, reducing initial microbial load so that less severe spoilage targets suffice, raising initial temperature by hot filling, and acidification where it is acceptable to the product.
Thermal processing causes some loss of water-soluble and heat-labile vitamins — thiamin, vitamin C and folate are the most sensitive — partly by destruction and partly by leaching into the liquid, which is often discarded. Fat-soluble vitamins, minerals, protein and fibre are largely retained, and some nutrients become more available: heat makes lycopene in tomatoes more bioavailable and destroys anti-nutritional factors such as trypsin inhibitors and lectins in legumes. Canned foods are thus nutritionally comparable to the cooked fresh food in many respects, though specific losses depend on product, process and storage time. Quality and nutrient retention can be predicted with the same kinetic models used for microbial lethality, using D and z values for the nutrient, which makes cook value and retention calculations a standard part of process optimisation.
Because underprocessing or container failure in a low-acid food can cause botulism, most countries regulate low-acid and acidified canned foods more closely than other processed foods. The international reference is the Codex Alimentarius Code of Hygienic Practice for Low-Acid and Acidified Low-Acid Canned Foods, which sets out requirements for raw materials, plant hygiene, establishment of scheduled processes, retort operation, container integrity, cooling water, records and handling of deviations. National laws build on it. In the United States, the Food and Drug Administration's regulations for thermally processed low-acid foods packaged in hermetically sealed containers (21 CFR Part 113) and for acidified foods (21 CFR Part 114), together with registration and process-filing requirements (21 CFR Part 108), are widely studied and are used as a model by exporters worldwide; meat and poultry products fall under the Department of Agriculture with similar rules. Many other jurisdictions, including India under the Food Safety and Standards Act regulations, the European Union under its general hygiene regulations, and others, have their own provisions; exporters must meet the rules of the importing country.
The central idea is the scheduled process: the process selected by a qualified person as adequate to achieve commercial sterility for a given product, container size and retort system, with all critical factors specified. Critical factors are any conditions that, if they vary, can affect the delivered lethality — for example initial temperature, fill weight, headspace, consistency or viscosity, particle size, salt or solids content, pH, rotation speed and retort venting schedule.
The scheduled process is set by a process authority: a person or organisation with expert knowledge of thermal processing and adequate facilities for heat penetration and temperature distribution work. The process authority designs the tests, analyses the data and specifies the process and its critical factors. In the United States, a commercial processor of low-acid or acidified foods must register its establishment with FDA and file the scheduled process for each product, container and process combination before producing it. Operators of retorts and aseptic systems, and container closure inspectors, must be supervised by someone who has completed an approved training course, commonly called a better process control school.
Every batch of a low-acid canned food must have records proving that the scheduled process was delivered: retort temperature charts or electronic records, reference thermometer readings, times of venting, come-up and processing, critical-factor measurements, container closure inspection results, and cooling-water treatment checks. Records are reviewed by a qualified person, usually within a short time after processing and before the product is released.
When a deviation occurs — the temperature drops below the scheduled value, a critical factor goes out of limit, a venting step is shortened — the affected product must be held. It may be reprocessed, destroyed, or released only after a process authority has evaluated the deviation, often by recalculating the delivered lethality with the general method using the actual temperatures, and has concluded in writing that the product is safe. If unsafe product has left the plant, it must be recalled. Codes printed on each container identify the plant, product, date and often the retort load, so that the scope of any deviation can be traced. This chain — validated process, controlled delivery, documented proof and expert evaluation of deviations — is the regulatory form of everything in the earlier chapters.
A student uses this subject to pass the thermal-processing calculations that appear in every food technology exam: D from survivor counts, z from two D values, lethal rates, F0 by the general method, Ball's process time with a given table value, and spoilage probability. In the pilot plant the same student fits thermocouples, runs a heat penetration test on a can or pouch, plots log(TR − T) to find fh and j, and integrates the cold-spot curve to check that a process meets its target. Knowing the pH 4.6 rule, the main spoilage types and the parts of a double seam answers most of the descriptive questions.
Five years into a cannery or ready-meal plant, a technologist runs retort and aseptic operations against scheduled processes, reviews batch records before release, carries out or supervises seam teardowns, investigates swollen or flat-sour returns by reading the spoilage pattern, and handles deviations: placing product on hold, recalculating delivered F0 from the actual retort chart with the general method, and sending the evidence to the process authority. They also support new products by organising heat penetration studies at worst-case fill, consistency and initial temperature, and by checking that labels, codes and container specifications match the filed process.
A senior thermal-processing manager or process authority designs heat penetration and temperature distribution programmes for whole product ranges, chooses between in-container and aseptic technology for a new plant, sets spoilage targets for markets with hot climates, signs off scheduled processes and deviation evaluations, and answers to regulators in audits and recalls. They weigh capital and energy cost against quality: whether agitating retorts, pouches or aseptic lines will repay their cost through better product, and how to standardise processes and records across plants in several countries with different rules.
A professor or research leader with long experience works on the open problems: validation of particulate aseptic processes, numerical and computational models of heating in complex containers, kinetics of spore inactivation including non-log-linear behaviour, combined processes such as pressure-assisted thermal sterilisation and microwave-assisted sterilisation, and quality optimisation. They advise regulators and standard-setters on whether new methods are adequate, train process authorities, and write the curricula and examinations that test this subject. They also judge the evidence behind proposals to change long-standing conventions — the D and z values used for target organisms, the 12D criterion, or the safety margins in deviation evaluations — where a wrong decision would affect every cannery that follows the standard.
This manual does not yet give the full Ball and Stumbo fh/U : g tables (questions supply the table value), the broken-heating-curve formulas, or the numerical solution of the heat conduction equation for finite cylinders. It does not give acceptable double-seam dimension ranges, which are set by can makers for each can size, or national incubation-test conditions in detail. Specific F0 values for commercial products are deliberately not listed, because they are set by process authorities for each product and container. Non-thermal and alternative thermal methods (high pressure, microwave-assisted and pressure-assisted thermal sterilisation) are covered in novel-nonthermal-processing; India-specific canning standards under FSSAI are covered only in outline.
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. Organism D values and regulatory limits quoted are standard textbook and reference values; real processes are set by a process authority for each product and container.
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.