Caballero B. (ed.) Encyclopaedia of Food Science, Food Technology and Nutrition. Ten-Volume Set

Подождите немного. Документ загружается.

operates by applying heat to turn the toast darker.

The toaster does not measure the color of the toast. It

does not correct for the fact that the toast is rye,

white, or sourdough, nor if the toast comes in differ-

ent thicknesses.

0012 Sensitivity to disturbance and the system’s inability

to correct for disturbances are the main disadvantage

of the open-loop system. Closed-loop control systems

may be used to overcome the drawbacks of open-loop

systems. A closed-loop control system uses an add-

itional measure of the actual output to compare the

actual output with the desired output response. The

measure of the output is called the feedback signal. A

simple closed-loop feedback control system is shown

in Figure 2b.

0013 A feedback control system uses a function of the

prescribed relationship between the output and the

reference input to control the process. At the com-

parison box (Figure 2b), the difference between the

output of the process under control (which arrives via

the feedback path) and the reference input is ampli-

fied and used to control the process so the difference

is continuously reduced. This difference is referred as

the input error (e(t) in Figure 2b).

0014 The comparison box generally consists of an

input transducer that converts the input signal to a

form used by the controller. At the measurement

box, an output transducer (sensor) is used to

measure the output response and also to convert this

response signal to the form used by the controller. For

example, if the controller uses electrical signals to

operate the valve of a temperature control system,

the input position and the output temperature are

converted to electrical signals. Potentiometers are

used to convert input position to voltage. A thermis-

tor, a device that changes its electrical resistance

with temperature, is used to convert output tempera-

ture to voltage.

0015The closed-loop system compensates for disturb-

ances by measuring the output response, feeding the

measurement back through a feedback path, and

comparing that response to the input at the compari-

son box (Figure 2b). If there is any difference between

the two responses (e(t) 6¼0), the system drives the

process to make a correction. If there is no difference

(e(t) ¼0), the system does not drive the process, since

the process response is already the desired response.

Closed-loop systems are also called feedback control

systems.

0016To illustrate the concept of feedback control,

Figure 3a shows a simplified manual control of a

baking process. On the outlet side of the oven, there

is an indicator (colorimeter) to provide the operator

with information on the actual value of the biscuit

outlet color (controlled variable). The operator

wants to keep the biscuit final color close to the

desired value. If the operator watches the colori-

meter and finds that the color value is higher (the

biscuit is darker) than the desired response, he/she

will increase the conveyor speed, i.e., the auger

r.p.m. It is considered in this example that the oven

temperature and air flow rate are maintained con-

stant for simplicity. The desired response, of course,

is in the operator’s mind and the operator makes all

of the control decisions. It is quite possible that the

color becomes lighter so the operator will have to

repeat the operation in the opposite direction until

the desired value is reached and kept constant during

the operation.

0017If the operator is replaced by an automatic control-

ler, as shown in Figure 3b, the system becomes an

automatic feedback control system. The actual outlet

biscuit color (the process output), measured by the

colorimeter, is compared to the desired response and

an acting error signal generated. The output color is

converted to the same units of the desired response by

Controller Process

Disturbance Disturbance

Output

(a)

(b)

Desired output

response

Comparison

e(t )

m(t )

Controller

Measurement

Process

Output

Desired output

response

fig0002 Figure 2 (a) Open-loop control system; (b) closed-loop control system.

3330 INSTRUMENTATION AND PROCESS CONTROL

a transducer. Based on the acting error value, the

controller calculates the changes that need to be

made in the auger r.p.m. to reduce and then eliminate

that error.

0018 Both manual and automatic controls operate

likewise. In this case, the operator’s eyes correspond

to the acting error device, the brain corresponds to

the automatic controller, and the operator’s muscle

represents the actuator.

0019 Advantages of closed-loop systems as compared

to open-loop systems include greater accuracy

and less sensitivity to noise and disturbances. The

transient response and the steady-state error can

be controlled easily and with great flexibility in

closed-loop systems. On the other hand, closed-loop

systems are more complex and expensive than open-

loop systems.

Computer-Controlled Systems

0020Since the 1970s, process controls have evolved from

pneumatic analog technology through electronic

analog technology to microprocessor-based controls.

0021Nowadays, many systems use a digital computer as

the controller. The use of a computer in the control-

system loop allows for the control and compensation

of many loops by the same computer through time

sharing. In addition, any adjustment of the controller

parameters can be made by changes in software

rather than hardware.

0022One important aspect of computer-controlled

systems is its real-time computing application. It

allows initialization of data acquisition opera-

tions and other tasks as control output calculations,

supervision, scheduling, and so on. Real time means

Raw

biscuit

(a)

(b)

Zone 1

Oven

Zone 2 Zone 3

Zone 1 Zone 2 Zone 3

Baked

biscuit

Baked

biscuit

Conveyor

Control auger

Auger-r.p.m.

Controller

Desired

response

Colorimeter

2.00

Color

Raw

biscuit

fig0003 Figure 3 Manual and automatic control of a baking process.

INSTRUMENTATION AND PROCESS CONTROL 3331

that an operation is to be completed within a specified

time constraint.

The Control Problem

0023 The control problem can be stated as follows: deter-

mine the one value of the manipulated variable

(input) that establishes a balance among all the influ-

ences (loads/disturbances) on the controlled variable

(output) and keeps this variable constant at the

desired value.

0024 The steps required to design a control system are:

process description, process and signal modeling,

design method, and control algorithm. Once the con-

trolled, manipulated, and load variables have been

identified, the next step is to describe mathematically

the process and signals (disturbances, reference vari-

ables, initial values). The design methods for linear

control systems can be classified into parameter

optimized control systems and structure optimized

control systems. Once the design method has been

selected, designing principles are applied to determine

the control parameters.

Closed-Control System Analysis and

Design

0025 From the control point of view, a system is composed

of controlled variables, manipulated variables (pro-

cess input), and disturbances. In a control scenario,

the number of manipulated (independent) variables

must be at least as large as the number of controll-

ed (dependent) variables. The controlled variables

should measure product quality directly or strongly

affect it. The manipulated variables should have

a large effect on the controlled variables (large

gain), should rapidly affect the controlled variables

(minimum delay, small time constant), and should

affect the controlled variables directly rather than

indirectly.

0026 Control systems are dynamic systems. That is, the

response of a control system to an input is generally

transient before reaching a steady-state response.

Consider, for example, the baking process

(Figure 3b) where a new biscuit is baked to a darker

color by changing the conveyor speed. When the

desired response (color of the biscuit) is changed to

a higher level (darker), the biscuit entering the oven

will leave the oven with a speed and color accuracy

designed for product quality. Figure 4 shows the input

and output for the baking system. The change in the

desired color value is the input and is represented by a

step command. Due to the limited power available

and to keep the products already in the oven from

breaking apart, it is not advised to have the oven

mimicking the suddenness of the input. The input

represents what the output must be after the conveyor

speed has changed; the conveyor follows the change

described by the curved region in Figure 4; this is

termed the oven’s conveyor response. This part of

the response is called the transient response.

0027After the transient response, the process ap-

proaches its steady-state response, i.e., approximates

the desired response. For the baking process example,

this response occurs when the biscuit reaches the

outlet side of the oven. The accuracy of the biscuit’s

final color is a second factor that could make the

output different from the input. This difference is

called steady-state error. Steady-state errors exist

when the control systems are defective.

Transient Response

0028In the baking process presented above, if the transient

response were too slow, the process would be ineffi-

cient (low throughput), whereas an excessively rapid

response would not be sufficient to bake the product

to the desired color. If the conveyor speed oscillated

around the desired response, the product color would

be different from one biscuit to the another, resulting

in poor quality. Therefore, transient response is very

important in the design and analysis of control

systems.

Steady-State Response

0029Another important objective in the analysis and

design of control systems is that the system has a

‘good’ steady-state error performance. The steady-

state response resembles the input and is usually

what remains after the transients have decayed to

zero. In the baking process, this response is the con-

veyor speed modified to a speed that produced the

desired biscuit final color.

Input command

Oven's conveyor response

Color

Time

Steady-state

response

Transient

response

Steady-state

error

fig0004Figure 4 Oven input and output.

3332 INSTRUMENTATION AND PROCESS CONTROL

Stability

0030 System stability under all system operating conditions

is another control system specification. A system is

unstable when its natural response (the way the

system dissipates/acquires energy independently of

the input) is much greater than the forced response

(dependent upon the input) making the system uncon-

trollable. This condition could lead to self-destruction

of the physical device if limit stops are not provided as

part of the design. Therefore, control systems must

be designed to be stable, i.e., their natural response

must decay to zero or oscillate as time approaches to

infinity.

Process Control

0031 The objective of a process control system is to main-

tain a variable in the process at its setpoint (desired

response). A process control system may be either

open-loop or closed-loop, but closed-loop systems

are more common. The process control industry has

developed controllers that are standard and flexible

for closed-loop systems.

0032 Examples of common features shared by many

process controllers include: (1) the values of the set-

point, process, and output variables are shown in

either analog or digital format; (2) the operator is

allowed to adjust the setpoint and switch between

automatic and manual control; (3) when manual con-

trol is selected, the operator is allowed to adjust the

controller output to vary the manipulated variable

in an open-loop control mode; (4) the operator

is allowed to adjust the control model settings to

‘tune the controller’ for optimum response; and (5)

to provide for remote setting of the setpoint by an

external signal. Many additional features are pro-

vided by microcontrollers (microprocessor-based

digital controllers).

0033 A control system (feedback or feedforward) per-

forms essentially three operations: (1) measurement;

(2) manipulation; and (3) signal transmission. Meas-

urement is an integral part of a control system. To

control a variable, its values need to be measured and

converted to suitable signals. Manipulation consists

of use of the control action to manipulate some vari-

able in the process in a way that will tend to reduce

the error. Signal transmission refers to signal commu-

nications between controllers and instruments.

Measuring Transmitter

0034 A measuring transmitter (sensor) senses the value of

a controlled variable and converts it into a usable

signal. Example of sensors include temperature trans-

mitter, flow transmitter, pressure transmitter, etc.

Consider, for example, a temperature transmitter.

The primary element of this sensor could be a thermo-

couple, a resistance element, a thermistor, or a filled

thermal element. The signal transducer receives the

output of the primary element and produces an elec-

tric current signal. In the case of a thermocouple, for

example, it converts temperature into a millivolt

signal, and the thermocouple transducer converts

the millivolt signal into an electric current.

Manipulating Element

0035A controller has two interfaces with the process it

controls: the input interface to the controller, and

the output interface from the controller. The input

side is handled by sensors and signal conditioners.

Various types of manipulating elements handle the

output side.

0036A manipulating element uses the controller output

to regulate the manipulated variable and usually con-

sists of two parts: actuator and final controlling elem-

ent. An actuator translates the controller output into

an action on the final control element, and the final

controlling element changes the value of the manipu-

lated variable. Manipulating elements include valves,

dampers, fans, pumps, heating elements, and electric

motors.

0037The pneumatic control valve is the most common

final control element used to manipulate the flow rate

of a fluid. Figure 5 illustrates a typical pneumatic

control valve. The input to the valve is air pressure

that acts on the diaphragm, compressing the spring,

thus pulling the stem out and finally opening the

valve. The actuator of a pneumatic valve can be

Air

Diaphragm

Actuator

spring

Stem

Plug

Seat

fig0005Figure 5 Pneumatic control valve.

INSTRUMENTATION AND PROCESS CONTROL 3333

described as direct actuator (if the actuator spring is

located below the diaphragm) or reverse actuator

(when the spring is above the diaphragm). The valve

position is the displacement plug from the fully seated

position. The valve action is in the direction in which

the valve plug moves as the air pressure increases.

If the valve closes as the pressure increases, an air-

to-close valve results. If the increased air pressure

opens the valve, it becomes an air-to-open valve.

Signal Transmission

0038 Most signals can be described as analog, digital, or

pulse. Analog signals typically vary smoothly and

continuously over time; digital signals are present at

discrete points in time. In most control applications,

analog signals range continuously over a specified

current or voltage range, such as 4–20 mA DC or

0–5 V DC. Digital signals are essentially on/off (the

pump is on or off), while analog signals represent

continuously variable entities such as temperatures,

pressures, or flow rates. Computer-based control

systems understand only discrete on/off information,

therefore conversion of analog signals to digital

representations is necessary.

0039 Transduction is the process of changing energy

from one form into another. Hence, a transducer is a

device that converts physical energy into an electrical

voltage or current signal for transmission. There are

many different forms of analog electrical transducers.

Examples of transducers include thermocouples and

resistance temperature detectors (RTDs) for measur-

ing temperature via voltage and resistance measure-

ment, respectively. Transmission channels may be

wires or coaxial cables. For noise-resistant transmis-

sion over significant distances, the raw transducer

signal is often converted to a 4–20 mA signal by a

two-wire, loop-powered transmitter.

0040 Industrial networks that transmit data using digital

signals are often an integral part of a process control

solution. Signals from digital instruments and con-

trollers are transmitted in digital format (on–off

pulses) using, for example, a number of parallel

wires. New instrumentation systems use a single

data highway to transmit digital signals to a series

of instruments and controllers. These data highways

are usually coaxial cables linked in serial or daisy-

chain (serial with complete loop). A microcomputer

built in the instrument and controller communicate

periodically over the highway by sending to or re-

questing from information between different devices

installed in the system.

0041 Examples of data highway are wide area network

(WAN) and local area network (LAN). Networks are

communication channels that connect large numbers

of stations to one or more central stations. WANs are

used to connect process systems separated by consid-

erable distance (branch offices in different cities).

LAN is a data highway used to connect local process

units with central operator displays and high-level

computers. LANs are used to connect process systems

located within a radius of about 1.5 km.

0042Nearly all digital network descriptions start with

the open systems interface (OSI) model. The OSI

model is based on several functional levels. The

lower levels are responsible for transmitting messages

between nodes on the same cable and from one cable

to another. The upper levels handle data formatting

and access security. A number of LANs are based on

the OSI model, such as the manufacturing automa-

tion protocol (MAP) and the Ethernet.

0043The MAP is a multilevel communication network

whose objective is to link together all the computers,

controllers, equipment, devices, and offices in an

entire factory. The Ethernet is able to handle error

detection and address the source of destination.

Control Modes

0044The operations performed by a feedback control

system include measurement, decision, and manipu-

lation. A closed-loop control system is shown in

Figure 2b. Measurement and manipulation have

been described above. Decision consists of calculating

the error (desired value minus measured value) and

using the error to form a control action.

0045The control modes convert the error into a control

action (controller output) that will reduce the error.

The most common control modes are the propor-

tional mode (P), the integral mode (I), and the deriva-

tive mode (D).

0046The simplest of the three modes is the proportional

mode. As the name indicates, the proportional mode

produces a control action that is proportional to

error. If the error is small, the control mode will

produce a small action. If the error is large, the con-

trol action will be large. To accomplish the propor-

tional mode, the error is multiplied by a gain

constant, K

. In a proportional control, the controller

output is algebraically proportional to the error input

signal to the controller as:

mðtÞ¼K

eðtÞð1Þ

where m(t) is the controller output, e(t) is the error

input signal, and K

is the controller gain. Eqn (1) is

called the control algorithm.

0047The integral mode produces a control action that

continues to increase its corrective effect as long as

the error persists. The integral mode increases slowly

when the error is small, and corrects more rapidly if

the error is large. Integral control action is based on

3334 INSTRUMENTATION AND PROCESS CONTROL

the principle that the controller output m(t) is propor-

tional to both the size and the duration of the actuat-

ing error signal e(t):

dmðtÞ

¼ K

eðtÞð2Þ

or:

mðtÞ¼K

eðtÞdt ð3Þ

where K

is an adjustable constant.

0048 The derivative mode produces a control action

that is proportional to the rate at which the error is

changing. If the error is increasing rapidly, the deriva-

tive mode will prevent the error becoming larger by

producing a correction action proportional to how

fast the error is changing. Derivative action (also

called rate control action) is based solely on the rate

of change of the error signal. An ideal derivative

action can be described as:

mðtÞ¼K

deðtÞ

ð4Þ

where K

is an adjustable constant.

0049 These three control modes can be used alone (with

the exception of the derivative mode) or combined.

The common control mode combinations are: P,

PI, PD, and PID. The proportional plus integral (PI)

control mode is used on processes with large load

changes when the proportional mode alone is not

capable of reducing the offset. The proportional

plus derivative (PD) control mode is used on pro-

cesses with sudden load changes. The proportional

plus integral plus derivative (PID) control mode is

used on processes with sudden, large load changes.

0050Figure 6 shows the typical response behavior of a

controlled process after a step change in the setpoint

occurs. The output variable is shown as a deviation

from the new setpoint change. If no feedback control

is implemented, the process slowly reaches the new

steady state. Proportional control speeds up the pro-

cess and reduces the offset. Adding integral action (PI)

eliminates the offset but causes the loop to oscillate.

The addition of derivative control (PID) made the

response faster and less oscillatory.

Advanced Control

0051The PID controller has been around for more than 50

years and is still today the most widely used industrial

controller. It is simple, easy to understand, and does

not require a precise process model to start up or

maintain. The PID works well if the process system

is simple. However, if the process dynamics change, it

needs to be retuned, and it has trouble controlling

complex systems, i.e., processes that are nonlinear,

time-variant, coupled, and have parameter or struc-

ture uncertainties.

0052Several advanced control strategies exist that can

provide improved process control over PIDs. Some of

these methods include cascade control, robust con-

trol, adaptive control, model predictive control, and

intelligent control. With the advance of computer

control, many of these techniques have been success-

fully applied in the food industry.

0053Cascade control involves two controllers, with the

output controller providing the setpoint of the

Time

No control

PID

Proportional

Offset

Setpoint

Output

fig0006 Figure 6 Step response of a feedback control system. PID, proportional plus derivative; PI, proportional plus integral.

INSTRUMENTATION AND PROCESS CONTROL 3335

secondary controller. The design of controllers or

control algorithms described up to now assumed

that only the control variable determines the process

input. This results in single control loops. By connect-

ing additional measurable variables to the single loop

(i.e., disturbances or auxiliary variables) it is possible

to obtain improved control behavior. These additions

to the single loop lead to interconnected control

systems.

0054 Robust control is a controller design method that

focuses on the reliability (robustness) of the control

algorithm. Robust control methods are suited to pro-

cesses that have large uncertainty ranges and small

stability margins.

0055 Adaptive control is defined as an intelligent feed-

back control system that adjusts its characteristics to

changes in the environment so as to operate in an

optimal manner. In process control applications, the

PID self-tuning scheme is widely implemented in

commercial products.

0056 Model predictive control (MPC) is one of the few

advanced control techniques used successfully in in-

dustrial control applications. The essence of MPC is

based on three key elements: (1) predictive model;

(2) optimization in range of a temporal window;

and (3) feedback correction. These steps are carried

on continuously by computer programs online.

0057 Intelligent methods in control are becoming part of

mainstream control approaches. They integrate con-

cepts and methods from areas such as control, identi-

fication, estimation, communication theory, and

artificial intelligence. Intelligent control paradigms

can endow control systems with functionality that

traditional methods could hardly make available.

They can also improve the analysis, design, oper-

ation, and maintenance characteristics of control

systems. Examples of this type of control strategy

include fuzzy logic and neural network.

0058 Fuzzy logic and neural network aim at mimicking

the operation of the human brain to some extent. A

control law described by fuzzy rules is more under-

standable by a human operator who knows the pro-

cess but who is not an expert control engineer. It

facilitates the integration of heuristic knowledge

about the system and theoretical control laws from

mathematical models, that are always an idealization

of the process.

0059Neural networks go further by aiming at building

models of the way a human being thinks and reaches

conclusions. In general, neural networks are useful

for representing and approximating nonlinear rela-

tionships. They are applicable for multivariable

systems and require less restrictive assumptions on

the process than conventional modeling and control

techniques.

See also: Plant Design: Basic Principles; Designing for

Hygienic Operation

Further Reading

Bateson RN (1999) Introduction to Control System Tech-

nology. Upper Saddle River, NJ: Prentice Hall.

Brescia L and Moreira RG (1997) Modeling and control of

a continuous frying process: II Control development.

Transactions of Chemical Engineering 75(c): 12–20.

Clarke DW, Mohtadi C and Tuffs PS (1987) Generalized

predictive control – part I. The basic algorithm. Auto-

matica 23(2): 137–148.

Clarke DW, Mohtadi C and Tuffs PS (1987) Generalized

predictive control – part II. Extensions and interpret-

ations. Automatica 23(2): 149–160.

Isermann R (1989) Digital Control Systems, vol. II. Sto-

chastic Control, Multivariable Control Adaptive Con-

trol, Applications. New York, NY: Spring-Verlag.

Moreira RG (2001) Automatic Control for Food Processing

Systems. Aspen.

Ogata K (1970) Modern Control Engineering. Englewood

Cliffs, NJ: Prentice-Hall.

Prett DM and Garcia CE (1988) Fundamental Process

Control. Boston, MA: Butterworth-Heinemann.

Schonauer S and Moreira RG (1995) Development of

a fixed-GPC controller for a food extruder based on

product quality attributes. Part I: control development,

implementation and analysis. Transactions of the Insti-

tute of Chemical Engineering Part C 73(C4): 200–210.

Seborg DE, Edgar TF and Mellichamp DA (1989) Process

Dynamics and Control. New York, NY: John Wiley.

Takahashi Y, Rabins MJ and Auslaner DM (1972) Control

and Dynamic System. Menlo Park, CA: Addison-Wesley.

Insulin See Diabetes Mellitus: Etiology; Chemical Pathology; Treatment and Management; Problems in

Treatment; Secondary Complications

Interesterification See Vegetable Oils: Types and Properties; Oil Production and Processing;

Composition and Analysis; Dietary Importance

3336 INSTRUMENTATION AND PROCESS CONTROL

INTERMEDIATE-MOISTURE FOODS

J Sych, Universite

Laval, Que

bec, Canada

Introduction

0001 Lowering of water activity (a

) to a level where

most microorganisms cannot grow is one of the

oldest methods of food preservation and forms

the basis for intermediate-moisture foods (IMF).

This article will review the nature and uses of

IMF by presenting the principles behind formula-

tion, some typical examples on the market, factors

limiting stability, and possible future developments

of IMF.

Principles Behind Formulation of IMF

0002 IMF have a

reduced to approximately 0.65 to 0.90

(i.e., moisture content of 15–40%), are plastic

enough to be consumed without rehydration, and

are stable without refrigeration. Other preservative

factors act in association with a

to insure the shelf-

life of IMF.

0003 a

is the amount of water available for deteriora-

tive processes in a food product and is defined as the

ratio of water vapor pressure of the product (p) to the

saturation pressure of water (p

) at the same tempera-

ture: a

¼p/p

. At low a

in food, at or below the

monolayer value, water is strongly bound to specific

sites and unavailable as a solvent for reaction. In the

range of IMF, the growth of microorganisms is

largely restricted, compared with high-moisture

foods, but certain reactions, such as Maillard and

oxidation, can still proceed, due to the increased solv-

ent and mobilization properties of water above the

monolayer. A reduction of a

is achieved by chemical

means (addition of humectants) and/or physical de-

hydration (evaporation, drying, ultrafiltration, etc.).

(See Water Activity: Principles and Measurement;

Effect on Food Stability.)

Processing Technology for IMF

0004 Traditional IMF are processed by the withdrawal of

water, by adsorption or desorption, and/or by the

addition of conventional humectants, such as salt

(sodium chloride) and sugar (sucrose). Although

some traditional IMF, such as dried fruits, are simply

dried to the desired a

, most IMF are prepared by a

combination of drying with added humectants.

Bakery products are prepared by adding water and

other ingredients to low-a

cereals, followed by

various processing and heating procedures to yield

in the range of IMF. (See Sucrose: Properties and

Determination.)

0005Novel IMF are often processed by using humec-

tants, such as glycerol, propylene glycol, and sorbitol,

for a

adjustment. Combinations of several humec-

tants, each at lower concentrations, are often used

to reduce off-flavor problems. Novel IMF may be

prepared by:

0006Moist infusion (desorption): solid food pieces are

equilibrated in a solution of lower a

, often with

heating. Drying may be used to reduce further a

(e.g., intermediate-moisture fruits)

0007Dry infusion (adsorption): solid food pieces are

dehydrated and then infused by soaking in a solu-

tion containing humectants (e.g., bite-sized cubes

of meat for space or military rations)

0008Blending: food components are weighed, blended,

cooked, and extruded using various processing

techniques (e.g., fermented meats)

0009IMF processed by adsorption normally have slower

microbial growth and lipid oxidation due to a lower

moisture content than IMF prepared by desorption

with identical a

. This is because of sorption hyster-

esis – the phenomenon in foods whereby the moisture

content of a food is lower during equilibration by

adsorption than by desorption. In spite of these

hysteresis effects and the waste of residual infusion

solutions, the desorption method is generally used

more, due to its advantages of lower cost, enzyme

inactivation, and better retention of flavor and struc-

tural integrity. With this method, it is possible to

reduce food weight by 50% by the simultaneous

countercurrent diffusion of water out of the food

and solute into the food.

0010In the production of novel IMF, pasteurization and

blanching are commonly used to reduce initial levels

of microorganisms and to inactivate enzymes. Anti-

mycotics, such as sorbic and benzoic acid and their

salts; methyl and propyl paraben; or calcium propi-

onate, assist in the control of mold and yeast growth.

Other chemical additives may also be incorporated to

slow down physical and chemical deterioration

during storage. (See Pasteurization: Principles; Spoil-

age: Chemical and Enzymatic Spoilage; Molds in

Spoilage; Yeasts in Spoilage.)

Effects of

on Microorganisms

0011Minimum a

for growth and toxin production of

some microorganisms capable of growing in IMF

INTERMEDIATE-MOISTURE FOODS 3337

are shown in Table 1. Generally, bacteria are less

tolerant of decreased a

than are yeasts, and yeasts

are less tolerant than molds (Figure 1). At a

< 0.95,

the growth of most Gram-negative bacteria, as well

as spore-forming bacteria, such as Bacillus and Clos-

tridium spp., is inhibited. Gram-positive bacteria,

such as Lactobacillus and Micrococcus spp., tolerate

lower a

and act as competitive microflora in fer-

mented intermediate-moisture (IM) meat. Among

the foodborne pathogens, Staphylococcus aureus has

the lowest minimum a

for growth (Table 1). Yeasts

and molds, referred to as xerophiles, are capable

of growth in high concentrations of solute, or at

 0.85. The control of bacterial growth in IMF

eliminates competition, which enhances mold and

yeast growth. (See Bacillus: Occurrence; Clostri-

dium: Occurrence of Clostridium perfringens; Spoil-

age: Bacterial Spoilage; Staphylococcus: Properties

and Occurrence.)

0012Reduced a

in foods affects the growth of bacteria,

yeasts, and molds by increasing the lag phase,

decreasing the rate of growth in the logarithmic

phase, as well as decreasing the maximum level of

development. Sodium chloride is the most effective

humectant on a mole-to-mole basis for reducing

, followed by glycerol, sorbitol, and sucrose.

However, at the same a

, microbial growth varies

with the nature of the a

-controlling humectant. For

relatively salt-sensitive bacteria, sodium chloride

exerts a greater inhibitory effect than polyhydric alco-

hols, such as glycerol, with sugars in between. In

contrast, for relatively salt-tolerant bacteria, such as

S. aureus, glycerol has a greater inhibitory effect than

sodium chloride. Many yeasts and molds are more

inhibited by sodium chloride than by sucrose, gly-

cerol, or glucose, at the same a

. The inhibitory

effects of sucrose and sodium chloride are attributed

to their ability to lower a

, whereas polyhydric

alcohols have less of an effect on a

, but appear to

cause damage to enzymes, and may have antimycotic

activity.

Moisture

Relative reaction rate

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0

Water activity

fig0001 Figure 1 Relative reaction rate as a function of water activity for foods. From Taoukis PS, Labuza TP, Saguy S (1997) Kinetics of food

deterioration and shelf-life prediction. In: Valentas KJ, Rotstein E, Singh RP (eds) Handbook of Food Engineering Practice, p. 381.

New York: CRC Press, with permission.

tbl0001 Table 1 Some microorganisms of concern in intermediate-

moisture foods

Microorganism Toxin produced Limiting a

Growth Toxin

Bacteria

Staphylococcus aureus Enterotoxin A, B, C 0.86 0.87–0.97

Halophilic bacteria 0.75

Yeasts

Saccharomyces rouxii 0.62–0.81

Debaryomyces hansenii 0.83–0.88

Molds

Aspergillus flavus Aflatoxin 0.78–0.80 0.83–0.87

A. chevalieri Xanthocillin X 0.65

A. ochraceus Ochratoxin 0.77–0.83 0.83–0.87

Penicillic acid 0.81 0.88

Penicillium viridicatum Ochratoxin 0.83 0.83–0.86

range of 0.65–0.90.

3338 INTERMEDIATE-MOISTURE FOODS

0013 For several genera of molds, higher a

is required

for spore formation than for germination of spores.

The time required for spore germination of micro-

organisms increases with reduced a

. Germination

of spores of some microorganisms occurs at a

values

below that required for growth, which suggests that

decreases in spore numbers can occur during storage

of some IMF. The type of solute also affects the

minimum a

for spore germination.

0014 The heat resistance of some bacteria and spores

increases under stress, such as reduced a

; this effect

is solute-specific and may be associated with the syn-

thesis of protective proteins. For heat-sensitive strains

of Salmonella, the increase in heat resistance can

be significant, particularly at high levels of small-

molecular-weight solutes. Such increases in heat

resistance must be taken into account in IMF produc-

tion, since heat treatments are often applied after

incorporating humectants.

0015 The minimum a

for toxin production of micro-

organisms is always higher or equal to that for growth

(Table 1). Mycotoxin-producing molds present the

greatest risk since many of their metabolites (e.g.,

aflatoxins) cause toxic effects in humans. Mold

growth and toxin production on IMF surfaces can

lead to penetration of toxins inside the product. In

addition, data on a

limits for mycotoxin production

in IMF are limited.

Other Preservative Factors of IMF

0016 Generally, water requirements of microorganisms for

growth and toxin production increase as environmen-

tal factors are changed from their optimum. IMF are

most often stabilized by reductions in a

in combin-

ation with other preservative factors, also referred to

as hurdles. Current views hold that the presence of

hurdles may interfere with the homeostasis or in-

ternal equilibrium of microorganisms and thus inhibit

their growth. A greater understanding of these prin-

ciples has led to benefits of improved product quality

and microbial safety; reduced additive levels; and

reduced refrigeration. Certain combinations of

hurdles can also result in additive or synergistic

effects allowing for a reduction of individual factors.

The most important hurdles used in varying inten-

sities to stabilize IMF are: available water (a

), heat

processing, acidification, redox potential, modified

atmospheres, preservatives, and competitive micro-

flora. Table 2 shows the a

of some IMF as well as

the other factors contributing to microbial stability.

0017 The combined effect of pH and reduced a

is used

in many foods to inhibit the growth of foodborne

pathogens (Table 2), as in IM meats where a pH of

less than 5 or a

of less than 0.85 is recommended

to inhibit S. aureus. The removal of oxygen during

packaging, i.e., low redox potential, inhibits obligate

aerobes (molds and some bacteria), and increases the

sensitivity of other microorganisms, including yeasts,

to a

decreases. This is used to delay rancidity and

mold growth in novel IMF, such as German mini-

salami with a

between 0.82 and 0.85. Protective

packaging is also an important hurdle used to main-

tain the stabilizing effects of redox potential and

modified atmospheres.

0018Many IMF contain added or natural chemical pre-

servatives to control the growth of molds and yeasts.

Legislation on the types of additives and their permit-

ted levels varies among countries. Although fre-

quently used, benzoic and sorbic acid require the

undissociated form (pH 4.5) for optimum anti-

microbial activity, which is often organoleptically

unacceptable in IMF. Another problem is the high

tolerance of some molds and yeasts to maximum

permissible levels of preservatives. Furthermore,

interactions between antimicrobials and a

may

exist; for example, Saccharomyces rouxii was found

to have greater resistance to sorbate at reduced a

The increasing interest in natural preservatives has

oriented much research towards the identification of

natural substances, such as those in garlic which

improve the stability of many traditional IM meats.

Edible coatings containing antimicrobials or antioxi-

dants can be applied to IMF to obtain specific func-

tional effects on food surfaces. For example, an edible

paste of ground garlic and other spices is used

to control surface mold growth during storage of

Turkish pastirma, a fermented IM meat.

0019Microstructure is also an important factor which

affects microbial growth, survival, and death in com-

minuted and blended IMF, such as fermented saus-

ages. The accumulation of natural flora as well as

tbl0002Table 2 Water activity (a

) and other preservative factors of

some intermediate-moisture foods

Product a

Preservative factors

Traditional IMF

Jams 0.82–0.90 a

, pH, F, Eh, pr

Dried fruits 0.60–0.75 a

, pH, F

,pr

Baked products 0.60–0.90 a

,F,pr

Meats 0.65–0.85 a

, pH, Eh, pr, cf

Novel IMF

Pet food 0.80–0.85 a

,F,pr

Ready-to-eat meat cubes 0.66–0.80 a

,F,pH

,pr

Fruit pastes  0.70 a

, pH, F, pr

Liquid tea concentrate 0.75–0.85 a

,pH

Fruit-filled cereal bars  0.85 a

, F, pH, pr

Liquid whole eggs  0.88 a

Possible preservative factor.

F, heat processing; pH, acidification; Eh, redox potential;

pr, preservatives; cf, competitive microflora.

INTERMEDIATE-MOISTURE FOODS 3339