Application of Surface Photo Charge Effect for Milk Quality Control


  • O. Ivanov,

    1. Author Ivanov is with Georgi Nadjakov Inst. of Solid State Physics, Bulgarian Academy of Sciences 72 Tzarigradsko Chaussee Blvd., 1784 Sofia, Bulgaria. Author Radanski is with Univ. of Forestry, Faculty of Veterinary Medicine, Dept. Infections Pathology, Hygiene, Technology and Control of Foods from Animal Origin, 10 Kliment Ohridsky Blvd., Corpus D, Office 309, 1756 Sofia, Bulgaria. Direct inquiries to author Ivanov (E-mail:
    Search for more papers by this author
  • S. Radanski

    1. Author Ivanov is with Georgi Nadjakov Inst. of Solid State Physics, Bulgarian Academy of Sciences 72 Tzarigradsko Chaussee Blvd., 1784 Sofia, Bulgaria. Author Radanski is with Univ. of Forestry, Faculty of Veterinary Medicine, Dept. Infections Pathology, Hygiene, Technology and Control of Foods from Animal Origin, 10 Kliment Ohridsky Blvd., Corpus D, Office 309, 1756 Sofia, Bulgaria. Direct inquiries to author Ivanov (E-mail:
    Search for more papers by this author


ABSTRACT:  The potential difference induced by the interaction of samples with electromagnetic radiation in the visible region is used for food characterization. In this article we show that the above effect can be applied for the understanding of specific reactions and processes taking place in milk such as change of the acidity and changes caused by an added reagent. We also propose a technique for instantaneous detection of inhibitors of starter bacteria in milk. We suggest possible methods for quality control of milk and other foods. Instantaneous results, practically no expenses for consumables, and possibilities for field measurements will be some of the advantages of this approach.


Since milk is of enormous importance as food source, supplement, or ingredient, a large number of analytical methods have been developed (Early 1998; Pomeranz and Meloan 2000) for its investigation. However, instantaneous tests, not requiring any consumables, are still in great demand.

The common qualitative or quantitative biochemical methods for milk analysis are based on various chemical reactions (Fox and McSweeney 1998; Wehr and Frank 2004). Such methods are Rose–Gottlieb method, Gerber method, Thorner method, Soxhlet–Henkel method, Kjeldahl method, and so on. For the range of standard compositional assays on liquid milk different reagents are needed, some of which are expensive, extremely unstable, and very toxic or cancer causing, such as isoamyl alcohol, sulfuric acid, ammonia, ether, formaldehyde, silver nitrate, trichloroethylene, and so on. As a rule, those methods are time consuming and requiring appropriate working conditions in laboratory environment, special equipment, and well-skilled personnel. Microbiological methods for milk purity testing (Roberts and Greenwood 2003; Jay and others 2005), serological methods for identification of impurities in milk, obtained after mixing different kinds of animal milks, fermentative, diffusive, and rapid test methods to prove the inhibitory substances in milk, have similar drawbacks (Schenck and Callery 1998; Hillerton and others 1999; Althaus and others 2003; Molina and others 2003; Linage and others 2007). These methods use nutritive mediums, sera, kits, strains of different microorganisms, and so on. Principle disadvantages of commonly utilized methods are related to their slowness, problems with consumables, and the need of laboratory conditions.

The present article reports on the application of an electric charge effect for new milk-quality control tests, providing easy and fast measurements. An appliance based on such an effect could be developed to be portable, operating without expensive consumables, providing fast results, and not requiring a laboratory environment.

Materials and Methods

The interaction of samples with electromagnetic radiation in the visible region results in generation of an alternating electric signal—Surface Photo Charge Effect (SPCE) (Borissov and others 1988; Pustovoit and others 1989). SPCE is shortly defined as follows: the interaction of every substance with electromagnetic field induces a measurable alternating potential difference between the irradiated sample and the common electric ground of the system, with the same frequency as the frequency of the incident field. The measurement is contactless and fast.

In contrast to other similar effects, the SPCE is present in any solid (Ivanov and others 1995), each solid generating a specific signal. For example, each body can be characterized by its weight, determined by the interaction between the body and the gravitational field of the Earth. In a similar way, it can be characterized by the SPCE, which is determined by the interaction of the body with an electromagnetic radiation.

When a constant electric field is applied, SPCE could not be observed. This peculiarity can be used for a simple and prompt verification whether the measured signal is due to the SPCE or to other similar effects, such as external and internal photoelectric effect, thermal electricity, and so on.

According to our experimental results, the SPCE is induced by electromagnetic field irradiation with frequency not only in the visible and the adjacent regions, but also for frequencies ranging from 1 Hz to 1 GHz. We suspect that, most probably, this effect exists in the entire electromagnetic spectral range.

At low electromagnetic field frequencies, the measurements could be performed at the emission frequency without additional modulation. In fact, the modulation is used only because it is still not possible to perform direct measurement of signals with frequencies in higher GHz range. In the present study, an electromagnetic radiation in the visible region (Terra Hertz range) was used, modulated with frequency of 800 Hz. An important feature of the SPCE is its significant dependence on the specific properties of the irradiated samples. This fact reveals opportunities for a rapid and contactless analysis, not only of solids, but also of liquids and gases (Ivanov 2006; Ivanov and others 2008).

A schematic diagram of the experimental set-up we used in our research is shown in Figure 1A. A continuous wave diode laser was used as a source (L) of incident radiation. The laser radiation power was I = 25 mW and its wavelength was λ= 655 nm. The incident light beam is chopped into periodic pulses using an optico-mechanical modulator (M) with a modulation frequency of 800 Hz. This value was chosen to be far from the grid frequency of 50 Hz where most parasitic signals could appear and, at the same time, to be in the working range of the optico-mechanical modulator. A pulsed laser or a pulsed LED also could be used instead of such a modulator. The sample studied was placed in the measuring arrangement (S). The latter represents a small vessel in which 2 electrodes are mounted. There exists also a possibility to use set-up with only a drop of liquid (Ivanov and Konstantinov 2002). The obtained signal is 20 dB amplified by the preamplifier (A). The signals measured are in the nano- and micro-volt scale. The detected signal had very low amplitude, thus a lock-in nanovoltmeter (N), capable of extracting the signal from the background noise, was utilized. The nanovoltmeter was coupled with a recording device to follow the time evolution of signals. The reference signal to the lock-in was supplied by the modulator (M). The configuration was described in detail elsewhere (Ivanov 2006).

Figure 1.

(A) Experimental set-up: L = light source; M = optico-mechanical modulator; S = measuring structure; A = high impedance amplifier with a gain 20 dB, input resistance of 108Ω; N = lock-in nanovoltmeter. (B) Sketch of the set-up for observation of SPCE: S = solid; I = solid–liquid interface, generating the signal; M = milk under study; E = electrode.

The measured signal is formed by the interface liquid–solid as discussed in Ivanov and Konstantinov (2002). The basic idea (Ivanov and Konstantinov 2000) is to put irradiated solid surface in contact with the milk under investigation. Since the electron properties of the solid surface are essentially influenced by the adjacent fluid layer, one can expect that optically excited changes in such a system will provoke measurable SPCE signals. In this way, with all other conditions fixed, it is possible to detect changes in the milk properties. The sensitivity of the method proposed is determined by the interface properties. It is expected to be sufficient, since the surface states are very sensitive to charge redistribution upon illumination in an appropriate spectral range.

This idea is illustrated in Figure 1B, where (S) is the illuminated solid (generally, it should be a substrate, for example, a semiconductor, generating an intense SPCE signal strongly dependent on changes in the fluid under study), (M) is the milk studied, forming an interface (I) with the solid in the spot of irradiation. The SPCE signal is measured by means of the electrode (E) using an appropriate apparatus.

Most probably, the measured signal was produced by the modulation of the double layer potential at the liquid–solid interface and, thus, of the measuring structure capacitance. Under certain conditions, this signal depends on the composition or other milk parameters and could be used for their investigation. As mentioned previously, we guess that SPCE in liquids is probably not an inherent feature of the liquid itself; rather it is generated at the liquid–solid interface.

For our study, we used raw cow milk from a herd mixed together from certified supplier and special care was taken to avoid sample contamination. Normally we used fresh, unprocessed milk, produced in the morning. It was cooled down to temperatures of 0 to 4 °C immediately after milking. Then the milk was stored in a cool place 4 to 6 °C for several hours until analyzed. The samples were heated back to room temperature just before measurement. Each milk sample had a volume of 10 mL.

Milk samples, prepared as described previously, were characterized by measuring the amplitude of the generated potential difference. We analyzed both the amplitude and its variation after the test liquid droplet of 0.04 mL was added to the milk. Such an amount was used since it was enough to incite reactions in all the experiments. The use of testing liquid droplets was aimed at initiating a specific response in the milk, registered by following the change of the SPCE-signal amplitude in a short period of time (20 s in our case). In this way we got the necessary information about the studied sample. The testing liquid was incorporated as in the incident spot of the laser beam so at a distance of 1 to 3 cm apart. The arrows in Figure 2 to 4 indicate the moment of time at which the testing liquid was added to the milk sample. In most experiments, adding of testing liquid provoked instantaneous changes in the measured signal, although in some cases there was a certain delay in the signal response. Here, by “instantaneous” response we mean that no visible delay was observed. At this stage of the research, we have not made any detailed measurements with an accuracy of the order of fractions of a second. Occasionally, a change of the phase of the signal, together with a variation in the amplitude, was also detected.

Figure 2—.

Effect of various testing liquids on the signal amplitude for 20 s: (A) antibiotic pharmazin (added in the incident spot of the laser beam); (B) water (added in the incident spot of the laser beam); (C) concentrated acetic acid (added 1 cm away from the incident spot of the laser beam); (D) 30% solution of H2O2 (added 1 cm away from the incident spot of the laser beam); (E) antibiotic tetravet (added in the incident spot of the laser beam).

Figure 3—.

Signal variation for 20 s after introduction of different amounts of concentrated acetic acid. The acid was added in the incident spot of the laser beam: (A) after the 1st amount of 0.04 mL was added; (B) after the 2nd amount of 0.04 mL was added; (C) after the 3rd amount of 0.04 mL was added. The milk was well stirred after each quantity acetic acid was added to achieve homogenization.

Figure 4—.

Influence of the antibiotic kanamycin on the signal amplitude for 20 s. The antibiotic was added in the incident spot of the laser beam: (A) milk with addition of H2O2 (0.4 mL of 30% solution of H2O2 was added to 200 mL of milk). The sample had been stored for 2 h 30 min before the tests; (B) pure milk.


Note that the SPCE can be applied for analysis of specific reactions and processes taking place in milk. Similar processes can be induced by adding of testing liquid to the samples. We carried out a large number of tests. The general conclusion was that the observed changes in the relaxation curve of the signal (that is, the curve, showing the change of the SPCE-signal induced by the droplet) was specific for each testing liquid. This is due to the fact that different testing liquids caused different reactions and processes in the milk. A few examples are presented in Figure 2A to 2E: the graphs show the change of the signal at 20 s after adding acetic acid (c), hydrogen peroxide (d), water (b), pharmazin (a) -antibiotic, and tetravet (e) -antibiotic. In this figure, the signal amplitude corresponding to the case of pure milk is 180 μV as measured after 20 dB amplification in the preamplifier A. Some of studied testing liquids, like water (b), spirits, and milk, do not provoke any significant variations in the measured signal, at least not in the amounts we used. This implies that the detected signal changes were not due to mechanical perturbation caused by adding a test liquid. All curves in Figure 2 to 4 are presented in their authentic mode as they have been recorded by the set up without any additional artwork. To compare the response amplitudes for different curves, we plot along the Y-ordinate the range of signals measured.

The presence of inhibitors, that is, substances that slow down the development of lactobacillus microorganisms and thus prevent fermentation in dairy products, is one of the most important problems in the dairy industry. We studied many types of antibiotics as testing liquids because they are inhibitors too. The 2 antibiotics (a) and (e) in Figure 2 are given to show that different antibiotics induced different signal variations. Since hydrogen peroxide, a substance with preserving properties, is also considered as an inhibitor, we studied its effect on the measured signal (curve d). The results described previously demonstrate the opportunity for measuring SPCE-signals of milk samples and also the fact that adding of different testing liquids (acetic acid, antibiotics, and hydrogen peroxide) provokes different changes in these signals. Next 2 figures (Figure 3 and 4) demonstrate the opportunity for detecting changes in milk due to the presence of extraneous substances.

We observed signal variations not only when the testing liquid was added to the sample in the incident spot of the laser beam, but also when adding it some distance apart. The shape of the relaxation curve, can be affected by the position of the spot, at which the testing liquid was added to the sample, as well as by the amount of the testing liquid, the incident spot of the laser beam over the investigated structure, the type of the latter, and so on. That is why it was necessary to keep the experimental conditions the same, varying only the type of testing liquid. On the other hand, the experimental conditions could be varied to obtain the clearest experimental results for a certain testing liquid. The measured signal was also affected by the introduction of an additional amount (a 2nd and a 3rd droplet of 0.04 mL) of the same testing liquid or a combination of different testing liquids (Figure 3 and 4). This dependence can be used to obtain more detailed information on the samples. The previously mentioned considerations show the potential of SPCE for milk characterization.

Figure 3 illustrates the dependence of the signal on the amount of concentrated acetic acid: after adding the 1st amount of 0.04 mL (curve a), after the 2nd droplet (curve b), and after the 3rd droplet (curve c). The signal amplitude of pure milk in this case is 145 μV. We see that a saturation effect was observed: the response became weaker and weaker and more and more time delayed. This was due to the fact that the acidity of the milk became too high and the processes taking place during its increase attained saturation. Curve (a), showing the signal response to the 1st addition, is different from curve (c) in Figure 2, because in the present experiment the acetic acid was dripped in the incident spot of the laser beam, while in the experiment, presented in Figure 2, the distance between the incident spot of the laser beam and the dripping point of the testing liquid was 1 cm. We also measured the signal change when the latter distance was 0.5 cm. In that case, 3 well-defined negative peaks were observed. They could be detected immediately after the testing liquid was added to the sample. Therefore, if we consider the dependence of the signal on the distance between the dripping point of the testing liquid and the incident spot of the laser beam, the signal response would vary from a broad band, without clearly separated peaks, (Figure 3A), to 3 separate peaks and, finally, to a single peak (Figure 2C) with much smaller bandwidth compared to the peak in Figure 3A. These results corroborate the previously mentioned conclusions regarding the dependence of the signal on the testing liquid characteristics and the experimental conditions.

The possibility for instantaneous tests for the presence of inhibitors in milk is significant for many practical applications. In this respect, it is important to study the effect of a testing liquid in samples of pure milk and in milk with an inhibitor added. Figure 4A and 4B show the effect of adding 0.04 mL of 25% solution kanamycin antibiotic on the signal amplitude of milk with hydrogen peroxide and pure milk, respectively. The initial signal amplitude before adding the kanamycin is 752 μV in Figure 4A and 162 μV in Figure 4B. The 2 curves are registered under the same experimental conditions; hence, the signal amplitude difference is due only to the difference of milk samples. The sample containing H2О2 is prepared as follows: a droplet of 0.4 mL of 30% solution of hydrogen peroxide is added to 200 mL of milk, followed by an efficient mixing. The so-prepared sample has been left untouched for 150 min before starting the investigation. Amounts of 10 mL are further taken for each experiment.

The difference between the 2 signals obtained is obvious. Based on this, methods can be developed for instantaneous tests and qualitative characterization of the presence of hydrogen peroxide in milk without any need of substantial expenses for consumables.

As another application option, it is worth mentioning that SPCE can be used for the development of new milk-quality tests, for example, for distinguishing milks obtained from different animals. Unprocessed milk produced by different species of animals, generated signals with different amplitudes. The following types of milk were analyzed: cow's, sheep's, goat's, and buffalo's. The amplitude of the signal generated by sheep's milk was reduced by 15% compared to the signal, generated by the cow's milk. The amplitude of the signal, generated by the buffalo's milk sample was 50% smaller than the one generated by the cow's milk. If this method is to be applied for analysis of goat's and cow's milk, the accuracy of the measurements must be further improved by increasing signal-to-noise ratio. For now it is an open opportunity. Prolonged studies are necessary including milk originated from different sources to estimate how much the effects of milk batch-to-batch variation will allow the direct control of milk species. It is always possible, of course, to develop indirect techniques like adding of testing liquid to improve the sensitivity of the method. At this stage, it is important that we have demonstrated that the SPCE-signal is sensitive to the different species of milk.


We would like to point out that the shape of the curve in Figure 4A depends on the period of time, during which the hydrogen peroxide was present in the milk, however, it always remained different from the curve in Figure 4B. This variation of the signal suggests the existence of continuous processes and changes in the milk due to the presence of hydrogen peroxide. The SPCE effect can be applied to study those processes as well. To detect the presence of an inhibitor in the milk in this case, it is not even necessary to record the signal. It will be enough just to compare the signal amplitudes of pure milk and the analyzed specimen since the signal amplitude of H2О2-containing milk is 4.64 times higher in that case. Or, if there is no pure milk sample available, one could follow the dependence of the signal with time and observe if there is any short-term change. The presence of such changes will be an indication for fast processes taking place in the sample caused by the external perturbing factor—hydrogen peroxide, in this case. The presence even at this initial stage of 3 different opportunities for detection of H2О2 in milk prompts that it is possible to develop different direct and indirect methods for each particular case. This could help in solving the problem of large batch–batch variation in the composition of milk under normal conditions. It should be clear, of course, that for the most of cases the development of highly sensitive techniques for eliminating the problem mentioned previously will need a long work.

Note that at the present stage of work, investigations are mostly qualitative ones, that is, they could help one to know whether there is or not a problem with the analyzed milk. They are not able to identify the admixture type and its quantity in the milk. Considerable study is necessary to make possible quantitative estimations. Nevertheless, a qualitative evaluation of milk purity could be very useful for practical applications, if accomplished in an express way, in real working environment, and without considerable expenses; these features are typical for the techniques proposed here.

Variations in the signal response, similar to those presented in Figure 4, were also observed in the presence of antibiotics in the milk samples. It is well known that antibiotics could be passed into milk if the cattle had been pretreated with antibiotics. However, we did not do any experiments with milk produced by animals pretreated with antibiotics because of financial difficulties.

The authors' arguments for ascribing the sensitivity of the observed signals to specific changes in milk, in particular to the presence of inhibitors, are summarized subsequently:

  • 1Samples prepared at the same time from the same milk source under same experimental conditions, but having different admixtures give rise to different signals specific for the admixture type.
  • 2Dripping of a 0.04 mL droplet of another liquid into the milk sample incites changes in the signal measured, which are specific for the additive used. In some cases, changes are not observed. This means that signal changes are not a result of mechanical perturbations and really are caused by reactions in the sample. This conclusion is supported by the fact that in some cases the change of measured signal appears not instantly but with considerable delays (Figure 3C).
  • 3Experiments show that the measured signal varies depending on the processes in milk. Examples for that are: temporal signal changes when another liquid is added to the milk; signal changes in unaffected milk left without any external perturbations at room temperature—in that case changes in the signal will be observed since milk left at room temperature will quite rapidly suffer bacterial degradation, which would result in a change in the signal; signal changes due to sudden change of milk's acidity, and so on. The comparison of temporal signal changes in pure milk with the ones in milk containing an inhibitor (for example, hydrogen peroxide) shows that they are substantially different.
  • 4The fact that the SPCE signal depends on the type of processes taking place in the milk shows that it is possible that appropriate experimental conditions could be arranged so that the presence of an inhibitor in the sample to be monitored.
  • 5This is demonstrated by our investigations in the case of hydrogen peroxide as an admixture. The required information can be achieved either by direct observation of the signal's amplitude or using appropriate reagents (testing liquids) modifying the signal.

As said previously, SPCE exists in every kind of solid under illumination. Most probably the presence of a liquid changes strongly the effect at the illuminated solid–milk interface and thus the capacity of the measuring structure. Correspondingly, any minor changes in the milk (concentration, contaminations, pretreatment, and so on) change the SPCE signal, indicating that the processes at the interface are strongly dependent on its properties. At this initial stage, the main achievement is the experimental evidence for the possibility to analyze milk in such a way.


SPCE can be used for development of new tehniques for milk tests. This method gives great possibilities because it combines optical probing of the sample with electrical detection of the generated signal. Instantaneous results and practically no expenses for consumables will be some of the advantages. Such devices will be cheap to produce and small enough to be used in field work. The use of a laser as a light source will not increase its size and weight since there now exist very small and compact laser modules even possessing internal intensity modulation.

Besides milk, SPCE can be applied for characterization of other foods (for example, bee honey) as well.


The authors thank Dr. Z. Peshev for useful discussions during the article preparation.