Dielectric Spectroscopy for Online Deep Frying Oil Monitoring

 

V.V. Meriakri* , S.V. von Gratowski,**

 

*LABORATORY OF SPECTROSCOPY AND MILLIMETER AND SUBMILLIMETER WAVE MEASUREMENTS, * ** of  Institute of Radio Engineering and Electronics, Russian Academy of Sciences (IRE RAS), Vvedenski sq.1, Fryazino, Moscow region, 141190 Russia, Tel: 7 (095) 5269266, Fax: 7 (095) 7029572, URL http://www.meriakri.de.vu , emails   * meriakri@ms.ire.rssi.ru ,  v-meriakri@gmx.net  ** svetlana@gratowski.de 

 

Fried food is very popular all around the world, widespread way of food preparation and comprises a wide variety of different products. Due to the high temperature during deep-frying process the fat deteritorates rapidly. The rate of degradation of fried oil depends on some parameters. Several factors, such as contact with air, temperature and duration of heating, degree of oil unsaturation, and presence of pro-oxidants or anti-oxidants, affect the overall performance of frying oils by diminishing their original characteristics. To these factor concern also type of fat, the fried food, and the some other frying conditions. During frying, a complex series of various chemical reactions and physical phase transitions take place, for example thermoxidation, hydrolysis, polymerization and fission and some others. At frying temperatures, oxidation can proceed rapidly. Excessive oxidation of oil and fats is often accompanied by polymerization. As oil and fats undergo heating in the deep-frying process, various decomposition products are formed. During frying, some components are produced which are considered to be harmful to human health, such as trans fatty acids, highly oxidized or polymerized constituents of fatty acids and acrylamide. [9] Frying is a dehydration process. When food is fried, water and any material and that water are pumped out of the food into the surrounding oil. Frying oil changes with use, going also from interaction between oil, water and food components. Thus it is not possible to suggest a general period of usage for deep frying oils. Determination of frying oil quality depends on many factors and is to be analyzed during frying sessions.

 

On the other hand the consumption of deep fried foodstuffs strongly increase every year, but oxidative stressed frying fats and oils are suspicious for health damages, some components are produced during frying which are considered to be harmful to human health, such as trans fatty acids, highly oxidized or polymerized constituents of fatty acids and acrylamide. Nowadays the production of healthy foods, as well as food with improved sensory characteristics is very important. That is why control and monitoring of deep frying fat is of great importance.  The safety of used frying oil depends on the presence of both polar material and polymer content. Nowadays there many physical methods linclude density, viscosity, smoke point, colour, refractive index, UV absorption, infrared spectroscopy and dielectric constant and chemical tests for the determination of free fatty acids (acid value), iodine value, anisidine index, saponification value, non-oxidised monomer fatty acids, polymerised triglycerides, petroleum ether insoluble fatty acids and total polar compounds. [4], [10], [1]

 

3rd International Symposium on Deep-Fat Frying worked out recommendations for Frying Oils analysis [3].

 

Analysis of suspect frying fats and oils should utilize two tests to confirm abuse. Recommended analyses should be:

 

1) Principle quality index for deep-fat frying should be sensory parameters of the food being fried

 

2) Analysis of suspect frying fats and oils should utilize two tests to confirm abuse. Recommended analyses should be:

1. Total Polar Materials (24%)
2. Polymeric Materials (12%).

 
3) The use of rapid tests for monitoring oil quality are recommended. Rapid tests should exhibit the following characteristics:

·       Correlate with internationally recognized standard methods,

• Provide an objective index,
• Be easy to use,
• Safe for use in food processing/preparation area,
• Quantify with oil degradation,
• Field rugged.

 

Other important and desired characteristics for rapid tests are:  [5, 2, 13, 14]

 

·       Not dependent from type of food and fat

·       Can be used in wide temperatute range

• Don“t be sensitive for water content.     

 

Conventional analytical methods for determination of the degradation of deep-frying fats are highly time-consuming and labor-intensive. In addition large volumes of solvents are considered as environmentally problematic. Several quick tests have been developed based on physical parameters (viscosity, dielectric changes(Food Oil Sensor (FOS))) and on chemical parameters (free fatty acids, oxidized and polar compounds). To cover the physical properties and chemical changes of frying fats many methods are proposed in literature. [3], [10]. [12], [13] The test kits consists of portable instruments or simple colour reaction sticks with a color. As is known, the dielectric constant is a good measure of the level of polar components in a frying oil. This has been proven in numerous investigations with the Food Oil Sensor (FOS) from the Northern Instruments Corporation. This has been proven in numerous investigations with the Food Oil Sensor (FOS) from the Northern Instruments Corporation. This test based on changes in dielectric for used oils.

 

Changes in dielectric for used oils.

These events are possible causes of major change (0.10 increase) in dielectric for used oil.
· High soot
· Oxidation
· Acid formation
· Water
· Glycol coolant
· Wrong or mixed oils.

 

Frying is a dehydration process. When food is fried, water and any material and that water are pumped out of the food into the surrounding oil. Hydrolysis is the reaction of food water with frying fat which leads to the formation of free fatty acids . The rate of hydrolysis or free fatty acid development depends on the following factors:

• Amount of water released into the frying fat (water is generally introduced from the food that is being fried).
• Frying temperature (the higher the temperature, the more rapid the rate of free fatty acid production).

 

In [6] was discussed papers which investigate the comparison between FOS and laboratory analysis methods, this was founded that moisture and particles in the oil samples can affect FOS readings.

 

 

All this show known disadvantages of the device are: [1] [5], [6]

 

• high purchasing cost
• necessary adjustment with reference substances
• insufficient assessment of the measured results by the device
• necessary cooling of the frying oil before measuring ( depend on temperature)
• strong influence of the water level of the oil
• necessary filtration for accurate results
• sensitivity in doubt ?

·       Depend on salt 

• depend on type of fat

 

These disadvantages are the reason for testing the technique of microwave and millimeter wave spectroscopy as a potential and alternative method for analyzing frying fats. In this report on line test for deep frying oil monitoring by means of microwave and millimeter wave  spectroscopy is proposed. Dielectric properties of many dielectrics, also many foods and oils depend on frequency. That is why dielectric spectroscopy give more information about dielectrics (food, oils) as a investigation of dielectric constant at zero frequency. With the use of this additional information it is possible to worked out rapid test or method for on line in-flow monitoring, what don’t have  disadvantages of tests which are available now. The method is low cost, highly accurate, non-destructive and easy in use.

 

Proposed method can solve also problems, which has FOS sensor. 

 

In the millimeter wave frequency range the  polarity of many polar materials reach their asymptotic values, and these asymptotic values for many oils almost don’t vary from each other. That is why in these frequency range it is possible to worked out a method, which don t depend on type of fat, it is not necessary adjustment with reference substances.

 

In the millimeter wave range dielectric properties don t depend on present on salt, metals and other conducting impurities.  So this method provide possibility for test, for which it is not necessary filtration for accurate results, and  which is not depend on salt, metals and so on, what present in used oil [11].

 

In microwabe and millimeter wave frequency ranges it is possible to worked out construction, which will be working on line and in flow. That is why it is not necessary cooling of the frying oil before measuring.

 

Water content has an great influence on dielectric properties in millimeter wave range. But this influence is very good known and in proposed method we plan determinate also water content [7]. After determination of water content influence of water will be eliminated from result of measurements.

 

The doubts in sensitivity of FOS sensor connected, first of all with the presence of water and other conducting impurities in used oils. In proposed method this problems will be solved.    

 

Other reasons for use of in line in flow monitoring for deep frying process. 

 

In production of deep frying food  as a in all food processing, the general rule is that the effective methods must be carefully applied to conserve the original qualities of the raw materials. Raw foodstuffs for deep frying have complex composition of different components. The balance of these components is a unique challenge of every year, of every delivery of raw foodstuffs. On-line monitoring gives possibility not only to check up the food quality, but also to correct and to optimize food production process depending on properties of initial food components.  That is why on-line monitoring under food production is of great importance.

 

This requires both monitoring of the raw components and permanent monitoring of deep frying production process and the quality of the final foodstuffs. Such methods are not affordable for process control in deep frying food processing industry for SMEs or individual producers. This aspect is decisive factor for many branches of food and beverage industries that are dominant for a large number of small producers. An adequate monitoring requires both low cost and simplicity of devices to be used.

 

Changes of physical and physical-chemical states of foodstuffs attribute deep frying processing and food shelf life. The key parameters for these phase transitions are time, temperature, polar substances content and polymers. Adequate monitoring of kinetics such phase transitions must have possibility to be carried out at all possible changes of main parameters of these transitions: time, temperature, etc. Thus this must be continuous on-line monitoring of physical state together with simultaneous monitoring of polar substances that influence on phase transition in wide temperature range. The phase transition like hydration, caramelization (brouning), polymerization, lipid oxidation in oils, is marked itself by a change of frequency dependent dielectric properties. Microwave, and millimeter wave spectroscopy appears to be an unique tool, working on line in wide temperature range, that is especially well adapted for continuous, food process monitoring due to its high sensitivity to polar components present and simultaneous for monitoring in changes of physical-chemical state in food products characterizing by dipole relaxation of substances.

 

In view of swift development of telecommunication market and information technology that have stable tendency to increase working frequencies of electromagnetic waves, as well as easy of realization, microwave and millimeter wave such sensors can be decreased to the values that are affordable not only for SMEs or individual producers but also for domestic consumers. Dielectric properties of substances are the measures of a food sample’s response to electromagnetic fields such as occur in a microwave oven. However, microwave radiation in such microwave sensors are 10000 times less than in microwave ovens and has no influence on the food quality.

 

Such sensors are non-existing on the market, although many research groups are working on methods and designing of such sensors.

 

Other background for the application of microwave spectroscopy for food industry    

There are currently some techniques for non-destructive on-line evaluation of the quality of foods bases on reflectance and absorption of ultrasonic and optical waves, NMR, what can be used for industry process control under food processing. But all these methods have serious limitations. NMR is highly sophisticated and expensive method, not affordable for SMEs or individual producers. Ultrasonic measurements are very complicate, acoustic wave have high attenuation in most foods; speed and reflectance of sound waves depend on multiple inside/outside factors.

Optical techniques are based on determination of refractive properties. Near and far infrared light transmittance techniques needed transparency that is not typical for the most of food substances. Limitations of near and far infrared reflectance and absorption caused by very small depth of penetration into a substance: there take place influence of many absorbing/ reflecting admixtures, surface irregularities, etc. Obtained experimental data show dependence of measuring results not only on food properties, but also on many outsides factors. These limitations make these methods not suitable for process control of real food processing industries. Microwave and millimeter wave-based techniques don’t have such limitations and can be used in optically opaque food substances and for both local and volume-averaged measurements, not only for surface ones. 

 

Microwave and MM wave monitoring have following advantages:

-        non-destructive, very low power of microwave radiation, there is no necessity to locate a sensor inside a food under investigations (advantage over HPLC, electrophoresis, DSC);

-        contactless; using non-contact sensors means not interfering with any workflow processes and causes no abrasion of the sensors or sticking of material (advantage over HPLC, electrophoresis, acoustic methods);

-        low cost (advantage over HPLC, electrophores);

-        fast (< 1 sec) and accurate (to less than 1% error) due to its high selectivity (advantage over HPLC, DSC)

-        high sensitivity due to high accuracy of electromagnetic measurements;  

-        negligible dependence on  outer factors (advantage over light, IR, FIR reflectance and absorption techniques);

-        very easy to operate (advantage over HPLC, electrophores, DSC);

-      compact size (advantage over HPLC, electrophores);

-      possibility of easy locating device at the right point in the process line (advantage over HPLC, electrophores, DSC);

-        possibility to give volumetric polar substance content, not only the surface one (like for light, IR, FIR reflectance and absorption techniques);

-        local and averaged monitoring (advantage over light, IR, FIR reflectance and absorption techniques);

-      determination of polar substance content independently from a materials weight, density and presence of nonpolar components and from temperature;

-      no pre-treatment of the samples (advantage over HPLC, electrophores, DSC);

-      easy to calibrate;

-      devices of long-term duration, they can be exploited for a long time without substitution of units  (advantage over HPLC, electrophores);

-      unification of units for devices;

-        measuring results can be transmitted to a computer for further data processing and statistical data analysis;

-        can be used in optically opaque or electrically conductive as well as non-conducting (dielectric) materials (advantage over  light, IR, FIR transmittance techniques);

-        no limitations because of attenuation in most foods (like for ultrasonic);

-        no limitations through hazard to personnel (like for X-ray);

-        need no unhealthy trade production of chemical reagents (advantage over HPLC, electrophores, DSC).

 

 

Application of microwave and millimeter wave-based methods and dielectric spectroscopy give key information about polar products’ content in admixtures of foodstuffs. They provide fast, on-line data about food production processes. These methods are unique, low cost ones applying on-line built-in sensors for simultaneous multiparameter process control of food processing industries.

 

On the other hand, microwave and millimeter wave-based measuring methods can be used not for all the food substances. These techniques can be used also jointly with measuring devices based on another physical principles (that is being a part of a complex measuring devices). These methods used in sensor fusion often include multivariate statistical techniques and artificial neural networks. In addition combining the data from several various sensors and transferring them into a more global quality parameter can also be used to increase the accuracy of several related measurements. In such sensor systems are used advantages of all the methods; they have minimal limitations.

 

 

                                                 

Determination of Total polar Materials and Polymeric Materials.

 

Not used oils are non-polar low-loss materials in microwave and millimeter wave frequency range. Their absorption features are of two types [19,21].

 

1.  A continuum absorption arising from vibration in the amorphous region. This absorption may extend into MM and microwave region too.

2.  An absorption due to impurities and degradation products.

Products of oil degradation are polar, they are not low loss materials.

Measurement method and circuits.

 

The relaxation frequency depend also on the size of molecules. Polar materials have much more smaller molecules then polymers, and compliance with this polymer material have lower relaxation frequency. That is why in this work on line in flow test for deep frying oil monitoring by means measurement dielectric properties on 2 different frequencies in microwave and millimeter wave range is proposed. With this method is it possible to measure simultaneously Total Polar Materials and Polymeric Materials at any temperature range, not depending of water, without pretreatment of oils.

 

For wavelengths l longer than approximately 6¸8 mm effective waveguide and resonator technique of low-loss materials properties measurement is elaborated [20,21]. On the other hand for wavelengths shorter than 0.5 mm very good Fourier transform and laser spectroscopy methods are available [15, 16, 21].

 

However, there are some difficulties in carrying out material investigation in the wavelength interval from 5¸4 mm to 0.6¸0.5 mm. The reason is that the waveguide technique is ineffective due to an decrease of the waveguide dimensions, gaps between waveguide and sample walls [22], on the other hand the optical technique is ineffective due to diffraction effect affecting the field structure and not allowing the use of geometrical laws of optics.

 

The diagrams of the experimental labor setups for measuring are shown on fig. 5,6.

 

 

 

 

The foto of labor setup is shown below.

 

 

 

For measurement of polymer material more low frequency microwave radiation was used

In this work all experiments was carried out on labor devices. For commercial using it is possible to use devices based on semiconductors and small size guiding structures. Such device can be built up inside deep-fryer or can measure non invasive through the small window inside deep-fryer.

 

 

Results

 

Sunflower oil (refined), Vegetable oil (refined (mixed)), Raps oil (refined) were investigated. Dielectric properties of these oils were investigated frequency range about 3 mm. The thickness of absorption layer was 4 cm. All these oils have the same values of dielectric properties: ε≈2.3 tanδ≈0.05 and dielectric losses 24 dB for 4 cm. From this results it is possible to see, that the results of measurements don t depend on type of oil. 

 

The most low-loss in millimeter wave range nonpolar liquids, for example (crude oil) have many times smaller dielectric losses. The results of our experiment show, that edible vegetable oil are weakly polar liquids. But polarity of  edible vegetable oil is very weak, because they have much more smaller tanδ as a polar liquids, that is about 0. 1 –1.0. [7]

 

The polarity of edible oils could result from, for example, present of fatty acids. 

 

 

Conclusion

 

1. This method can be used for on line in flow monitoring for used deep-frying oils. The method is non destructive, easy in use, does not depend on type of oil, salt and other impurity content.  

 

3.  This results  must be verified with any reference methods. That is why we are looking for collaboration with food and oil scientists to continue this work

 

References.

 

[1] Electronic Nose for Detection the Deterioration of Frying Fat —

Comparative Studies for a New Quick Test. Ulrich Demisch, Mike Muhl, Testo GmbH & Co, Lenzkirch, Germany 3 rd International Symposium on Deep Fat Frying. March 20-21, 2000, Hagen/Westphalia, Germany. p.11.

 

[2] Application of near infrared spectroscopy (NIRS) to the analysis of

frying fats H. Büning-Pfaue, S. Kehraus, Bonn. 3 rd International Symposium on Deep Fat Frying. March 20-21, 2000, Hagen/Westphalia, Germany.

[3] 3rd International Symposium on Deep-Fat Frying: March 20-21, 2000, Hagen/Westphalia, Germany. http://www.dgfett.de/material/recomm.htm  

[4]Chemical and Physical Parameters as a Quality Indicator of Used Frying Fats. Christian Gertz.  3 rd International Symposium on Deep Fat Frying March 20-21, 2000, Hagen/Westphalia, Germany.

 

[5] Veränderungen von Fetten und Ölen beim Erhitzen und bei der Lagerung Dr. Christian Gertz http://www.dgfett.de/material/fettverderb.pdf

 

[6] Van der Pals, Ernst, "The Measurement of Dielectric Constant as a Method for Quality Assessment of Frying Oil", 11th Scandinavian Symposium on Lipids,  Lipidforum Goeteborg, p.197 (1981).

[7] V.V.Meriakri, E.E.Chigrai, M.P. Parkhomenko, "Millimeter waves for water content monitoring in materials and media", 11 Feuchtetag 2002, Vortrage, Weimar, Germany, 18/19 Sept. 2002, pp. 13-22.

 

 [9] Oxidative stressed frying fats and oils.  Potential role for health Nikolaos K. Andrikopoulos. 4rd International Symposium on Deep-Fat Frying: 11-13 January 2004, Hagen/Westphalia, Germany

 

[10]Adsorbent Treatment of Frying Adsorbent Treatment of Frying Oil and the Impact on Health and Oil and the Impact on Health and Nutrition. Brian S. Cooke. 4rd International Symposium on Deep-Fat Frying: 11-13 January 2004, Hagen/Westphalia, Germany

 

[11] New Theoretical and Practical aspects about Frying Aspects about Frying Process. S. S. Parkash Kochhar Christian Gertz.  4rd International Symposium on Deep-Fat Frying: 11-13 January 2004, Hagen/Westphalia, Germany

 

[12] Developments in Oils Developments in Oils for Commercial Frying for Commercial Frying J. Barry Rossell J. 4rd International Symposium on Deep-Fat Frying: 11-13 January 2004, Hagen/Westphalia, Germany

 

[13] TESTS TO MONITOR THE QUALITY OF DEEP- FRYING FATS & OILS. Richard F. Stier4rd International Symposium on Deep-Fat Frying: 11-13 January 2004, Hagen/Westphalia, Germany

 

[14] FRYING AS A SCIENCE Richard F. Stier4rd International Symposium on Deep-Fat Frying: 11-13 January 2004, Hagen/Westphalia, Germany

 

[15] K.J. Button (edd.), Infrared and Millimeter Waves, Academic Press, N.-Y., Volumes 1-14, 1984.

[16] G.Chantry, Submillimetre Spectroscopy, Acad. Press, London-N.-Y., 1971.

[17] V.N. Apletalin, V.V. Meriakri, E.E. Chigrai, Quasi-Optical Techniques of Stadying Liquid and Solid Dielectrics at Submillimeter Wavelengths, Proc. Symposium on Submillimeter Waves, N.-Y., USA, pp. 631-641, April 1970.

[18] V.V. Meriakri, V.N. Apletalin, A.N. Kopnin et. al., Submillimeter Beam Wavequide Spectroscopy and Its [19]Applications, in book Problems of Modern Radio Engineering and Electronics, edd. V.A. Kotelnikov, Nauka Publishers, Moscow, pp. 179-197, 1985.

[20] A.C. Lynch, S. Ayers, Measurements of Small Dielectric Loss at Microwave Frequencies, Proc. IEEE, vol. 119, No 6, pp. 767-770, 1972.

[21] M.N. Afsar, J.R. Birch, R.N. Clarce, The measurement of the Properties of Materials, Proc. IEEE, vol. 74, No 1, pp. 183-199, 1986.

[22] V.V. Meriakri, About Errors of Wavequide Method of Dielectric Properties Measurements, Metrology, No 4, pp. 67-70, 1973, (in Russian).

[23] V.V. Meriakri, I.P. Nikitin, Iris Effects in Quasu-Optical Measurements of Dielectrics, in book Quasi-Optical technology of MM and SMM Waves ranges Kharkov, Ukraine, pp. 55-58, 1989, (in Russian).