A new generation of sophisticated analytical sensors is beginning to allow real-time on-line information on the plant floor of food manufacturing operations. There are a number of different ways to classify and describe on-line sensors for processing systems. Sensors may be invasive or non-invasive. Due to the harshness of food processing environments, it is always desirable to use sensors that do not contact the food system. However, in many situations including some of those involved in measurements of food ingredient properties or chemical reactions of food components, currently available sensors must come into contact with the food material .
In choosing which sensor to use for a particular application, selectivity, sensitivity, or cost may not be the only factors which must be taken into consideration. The sensor technology, whether optically, electrically, or acoustically based, is also a factor . In certain processing systems, for example, signal interference would preclude the use of an electrically-based sensor. If a sensor was to be used in a microwave oven environment to measure moisture, the components of an electrically-based system would be unacceptable. Even with an optically-based sensor, all measuring components within the oven cavity must contain no metallic components, including the protective sleeves used to shield the fragile optical fibers.
Sensor selection can also be made based on the active material used in the sensor and how it will react with the food materials in the system within specific temperature ranges. For example, in a high temperature baking process, ceramic materials may be a better choice for use in making moisture sensors than polymer-based materials since they can withstand much higher temperatures without being damaged.
Physical sensors and analytical sensors will measure different food characteristics and may, therefore, be used under different circumstances. For example, if the total weight of a product is the only measurement of interest, a classic load cell would be an appropriate choice. However, if the ratio of solid to liquid were desired, an ultrasonic sensor or one of the emerging microwave sensor technologies would be more appropriate. Kress-Rogers  illustrates these differing types of measurement capabilities and sampling techniques.
Sensor selection will also be based on consideration of the location in the process a measurement is needed. On-line measurements can be made at varying points during a process as shown in Figure 1. To select the ideal sensor for a given use, therefore, one must consider the selectivity for the property to be measured, the process operating conditions including the potential for electromagnetic interference caused by the process, the temperature, and potential interfering chemical species.
Components Moisture pH
Size Moisture Color Microbiological Environmental/Pesticides
Components Moisture pH
Size Moisture Flavor/Aroma Microbiological
Package Integrity Microbiological Shape Weight Temperature Pressure Moisture
Figure 1. Important applications for sensors in a generic food manufacturing operation
The various sensing technologies in use today correlate a specific property or molecular signature of the measured system with a fundamental property of the sensing technology or sensing material . For example, if one were measuring the amount of moisture in a food process with a moisture absorbing thin polymer film, electrical impedance of the film might be measured with no water present as a base line. Moisture absorbed by the polymer would change its impedance. A good sensing material would come close to showing a linear change with the moisture. Such new materials are emerging from research laboratories. Many of these are not yet patented. A great need for further understanding of new sensor materials remains because the sensing properties are extremely dependent upon their surrounding environment. For instance, some good polymer-based moisture sensing materials are seriously affected by other chemicals, particularly alcohols, and they can temporarily or permanently lose their measuring capabilities in the presence of such chemicals.
Optically-based sensing was identified at the beginning of the nineteenth century. The scientific principle has taken a long time to be adapted to a viable sensing technology. Considerable information about the optical properties of the measured systems is required to develop such technologies; much of this is still the subject of research studies . Optical spectroscopy operates according to the following principle, schematically illustrated in Figure 2. A light source is focused into a fiber optic cable which transmits the light with very little loss of strength to a sensor. The sensor interacts with the environment or food system to be measured either physically by absorption, or chemically with altered properties. This change in the sensor usually absorbs and reduces the light signal. After continuing through the fiber to a detector, the incoming and detected signals are compared and the change in signal correlated with the attribute environment being measured.
Optical sensing technology became recognized as a valuable measurement tool in the flour, grain and forage industry in the mid 1960's. Its background and principles are well described by Scotter . Improved spectral discrimination of instrument output as well as the development of new optical performing materials have led to on-line application in determining fat, protein, alcohol and moisture [4, 13]. The technology has continued to develop and more specific capabilities have begun to emerge in practical sensing systems [14, 15].
For example, specificity of optical sensing has been expanded using the electromagnetic spectrum from the visible light range, to near infrared, to far infrared. Also, new optical materials have been developed that reduce signal loss and extend the useful signal transmission range making on-line use very attractive. For example, the Fiber Optic Materials Research Program at Rutgers University has developed porous fiber optic materials which can be used in conjunction with specific chemically sensitive compounds to sense a range of properties including moisture, pH, NOx, S02, and H2S [14, 16, 17]. Others have investigated temperature and thermal profiling . Optical sensing technologies
Fiber Optic Cable
Figure 2. A schematic diagram of optical sensing
are being developed to meet needs in the processing of cereals, breads, cookies, dairy products, and fruits, as well as to measure properties of flour and other ingredients [13, 19-21].
One of the more significant areas of development that has aided in new use of optical sensing is the source of light. The light emitting diode (LED) and laser diode (LD) have made it possible to obtain much stronger light signals in narrow frequency ranges. This has helped overcome selectivity and signal to noise issues which affect the ability to effectively correlate measurements.
Several other optical sensing technologies should be noted. Using optical sensing, the fluorescence properties of certain materials can be exploited. Online moisture sensing developments using a polymer based fluorescent film have been reported by Pedersen, et. al. .
Color monitoring with infrared technology and color machine vision also have new on-line quality measurement capabilities to determine, for example, the effects of thermal processing on peas and carrots [15, 23]. Machine vision provides new techniques for inspection and placement of packaged food placeables, and certain inspections for product packaging .
Electrically-based sensing often uses ceramic or polymer materials that specifically correlate emf (voltage), impedance, or dielectric constant with the desired measurement property. Such sensing operates according to the following principle, shown schematically in Figure 3. A sensor, placed into an electrical circuit, reacts with moisture, heat, pressure, or some other property in the food environment, or by chemical reaction with the food system, causing a perturbation in the electrical properties of the sensing material. This change can be measured and correlated with the attribute of interest.
There are three major types of electrochemical sensors being used for process measurement: ion-selective electrodes (ISE), ion-selective field effect (ISFET) transistors and metal oxide gas sensors . The most commonly used ion-selective electrode is the glass pH sensor, an off-line device. This device cannot be used on-line in food processing operations due to its fragile glass structure .
To overcome this problem, ISFET sensors became important. The development of an ISFET sensor, particularly for pH measurement, was aimed at overcoming the hazard of the fragile membrane of the ion-selective electrode. Various insulating oxide films such as Si02 and A1203 have been investigated to optimize performance of the ISFET over a wide range of processing conditions since this type of sensor is influenced by a variety of ionic materials (e.g., Na+, Ca+), and temperature. The drawback to these for on-line use is that the adhesives used in their manufacture generally do not withstand the higher temperatures used in some food manufacturing processes ,
Metal oxide gas sensors may hold the most promise for use in high temperature food processing applications. They are widely used for measuring ethanol during fermentation processes . These sensors react with specific chemicals in the environment resulting in a change in conductivity or impedance. The complexity of most food systems makes understanding ion selectivity key for use of such sensors. Again, the research community continues to investigate improved materials for potential applications to other gases that may be related to product formulation (e.g., butanol, acetone, formaldehyde, etc.) [29, 30].
Polymer-based sensors represent another major class of electrically-based on-line measurement tools. Recent developments have given rise to a variety of commercial sensors based on the measurement of conductivity or impedance of a polymer film as it reacts with the food processing system environment. Early applications of polymer sensors were developed using piezoelectric materials to measure force . Both piezoelectric and other thin film polymers are under study for use in higher temperature on-line moisture measurement [32, 33],
One area where thin film polymers have recently begun to see utilization is in the electronic nose. Using arrays of sensors with either a lipid analogue or a tin oxide membrane supported on a polymer base, the response pattern to electric potential is determined. Multivariant analysis and statistical evaluations with techniques like the chemometric technique developed at the Center for Process Analytical Chemistry are applied to develop and correlate pattern recognition [34, 35], The concept of using flavor/odor analysis for product quality determination has great appeal. Currently, research is occurring toward discrimination of coffee blends and roasts. Considerable application development is still required. Advanced sensors in this area to measure flavor/aroma would find many opportunities for use including peanut roasting and snack foods.
The use of microwave sensing to date has been primarily associated with bulk moisture determination . Dielectric constant and loss factor are properties of food materials ideally suited for microwave determination of moisture content. Since, microwave instruments are dependent on sample density, their broad application is limited due to the difficulty of measuring density on-line. More recently, a new two-variable technique has permitted development of a more compact and inexpensive instrument. In addition, onboard microprocessors and data storage have enhanced data acquisition and processing capabilities [36, 37], A better understanding of food dielectric properties coupled with microwave advances provides opportunities for further on-line exploitation. The principle advantage of this instrument is its non-intrusive capability. The cost/benefit ratio may still be too high relative to other developing moisture sensors to see utilization broadly increased.
Nuclear magnetic resonance (NMR) must be mentioned as a potential online process sensing tool. While NMR has been extensively used in the laboratory to study the relaxation time of food matrices leading to determination of bound and unbound water, its cost is prohibitive for on-line use. However, new magnet designs that are far less costly as well as development of innovative food transfer systems through the magnet are under study [1, 3, 38], Chapter Two of this book focuses on the use of NMR for on-line quality assurance.
The use of thermal diffusivity/thermal conductivity to correlate with food moisture content has been developed into a new sensor for highly viscous materials such as doughs, cheeses, and emulsions and for measurements in powder materials. The sensing system, measuring the rate of heat absorption to the food sample, provides a correlation with moisture content and thermal conductivity of the sample from its calibrated database [39-42].
Another sensor technology receiving increasing attention and investigation is ultrasonics. Ultrasonic measurement in food systems is of high interest, like NMR, because of its non-intrusive nature. Ultrasonic technology is well established for determining fill levels and for in-process metering and has been adapted to determination of solids content in simple solutions. These sensing techniques use the principle of determining sound absorption, reflection, and phase shift to correlate with properties of the system attenuation . As more complex food systems are investigated, particularly nonhomogeneous media such as doughs or cheeses, signal interpretation becomes limiting. Nevertheless, a number of applications of ultrasonic technology to emulsions, milk, temperature measurement, and determining the quality of fresh products, have been reported [44-47]. Chapter Four of this book provides more information about ultrasonics.
2.2. Sensing methods
Clearly, beginning with the receiving and storage of ingredients, through blending, processing and packaging, measurements of temperature, pressure, weight (continuous and discrete), size, shape, fluid flow of gas and liquids, viscosity, and liquid level, are important for process and product quality assurance. Most of these sensors are currently in use by the food industry. New developments in computer-automated control will assist in fine tuning their use , In addition, food industry application of on-line rheometry for viscous materials like doughs is also emerging [49-51].
Chemical sensors are used to measure a variety of chemical and biochemical components in food systems. Physicochemical sensors are emerging as important to address the measurements of water (bulk moisture, gaseous moisture, and humidity); fat; protein; carbohydrate; chemical composition; soluble solids; and color. Measurement of some combination of these parameters often opens up a new control strategy toward optimizing product quality.
Companion to physical measurements, there are number of chemical parameters of significant influence to food quality. Their measurement will provide the specificity for good control of product manufacture. Chemical measurements include pH, ionic species (e.g., chloride), odors, flavors, dissolved oxygen, developed gases, and alcohols, as well as measures of freshness and ripeness. While most of these parameters can be measured off-line, new on-line sensors are emerging .
Biosensors are often defined as sensors which use a biologically sensitive material producing a biochemical signal that is converted by a transducer into an electrical response. The biomaterial is generally placed on or into a membrane. An important advantage of a biosensor is selectivity and the capability to detect very small amounts of such compounds as glucose. Sterilization, or exposure to harsh conditions in a food process are often fatal to a biosensor. Enzyme-based biosensors used in food measurements generate signals from chemical reactions. Immunosensors are based on antigen-antibody reactions. Currently, the main application of these latter biosensors is in the pharmaceutical industry. Enzyme-based biosensors are commercially available; glucose and lactate are the two main analytes for which instruments are available and routinely used .
Rapid microbiological determination increases production efficiency by reducing the time and interim storage needed until food can be considered safe enough for distribution. Again, off-line technology is capable and accurate but inadequate for the increasing throughput of modern food processing. Rapid measurement is needed, including bacteria, mold, and pesticides .
Was this article helpful?