Investigations for Development of Structures for Control of Fog Contaminations

. Characterization of fogs is an important problem that does not have a good solution in real conditions. To investigate this problem, we chose the electromagnetic echo effect (EMEE) because it offers several advantages - measurements are fast, accurate and can be contactless. Various concepts for the detection of impurities in the composition of fogs by implementing the EMEE were considered. An overview was made of the advantages and disadvantages of the proposed methods. An experimental setup was designed to conduct studies regarding the development of sensors for the control of fog contaminations. It has been experimentally proven that a change in the generated EMEE signal is observed when a change in the composition of the fog is present. The signal is also influenced by various conditions, e


Introduction
Fog is a natural phenomenon that people observe and study [1]. Fog is a collection of microscopic droplets dispersed in the air, and it is a dynamic structure that changes continuously in every part of its volume. Fog can contain many different impurities, for example, small particles, dust, various chemical substances, etc. In addition to being a natural phenomenon, fog is also used for many technological purposes -spreading substances, cleaning, air conditioning, ventilation, etc. These applications make the study of fogs and the creation of effective sensors for them very essential.
Different fog control sensors exist. One of the methods is to evaluate the visibility in the air, and this is most often realized with cameras [2][3][4]. The presence of fog (and/or other pollutants) leads to reduced visibility. A more common option is by using a light source (from UV to IR) and a light detector located at a certain distance from the source. When particles or droplets are present in the air, the intensity of light from the source decreases due to the scattering of light by them. LIDARs are also used to study the presence of fog in the environment [5]. Sometimes humidity sensors are used as well to detect fog. Radars can be successfully used to estimate fog density [6].
Existing methods either do not meet the requirements for controlling fog impurities or are complex, expensive and slow. Several problems exist in building an efficient fog impurity sensor. A major problem is that fog condensation on a solid-state detector distorts the results. In addition, the dynamism of the fog and the continuous change of its volume leads to a change in the amount of fog in contact with the sensor and incorrect reading of the results. Another difficulty is that the composition control sensor generates a signal mainly from the main substance of the fog, which is many times more than the impurity, and this greatly reduces the sensitivity of the impurity measurement, since the signal from the main substance will be much stronger.

Experimental setup
The research of our team has led to a series of new experimental results, revealing the so-called electromagnetic echo effect (EMEE). The effect is based on the interaction of any solid body with an electromagnetic field, which generates in the body an alternating, electric signal with the same frequency as the frequency of the incident field [7]. The EMEE signal is highly sensitive even to small changes in the composition and properties of the object under studygas, liquid, or solid. The method is fast and contactless, provides real-time analysis, and can be considered universal [8]. That is why we created an experimental setup to perform investigations for the development of structures for fog contamination control (Fig. 1).

Fig. 1
Simple representation of the experimental examination setup for the development of sensors for fog composition variation control The examined structure (1) can be irradiated by solid-state laser modules (2) with different wavelengths. Different wavelengths are needed to optimize the measurements using the examined structure. The laser modules have the ability to modulate the generated radiation in the range of up to 150 kHz, which at the moment is sufficient for our purposes. The proper supply of the lasers is ensured by the direct current power supply (3). The emitted radiation is modulated by an impulse generator (4). This generator also provides a reference signal for a lock-in nanovoltmeter (5), which measures the alternating electrical signals generated from the structure (1). The impulse generator allows for the examination of light-interface interaction at different modulation frequencies. The intensity of the laser beam can be regulated with a polarizing filter (6). This also allows for tweaking of the output signal. When necessary, the structure (1) is heated by a hot plate (7). The presence of fog is appropriately imitated by a fog generator (8).
Experiments were conducted with large variations of the examined structure (1). They can be classified into three main groups, which correspond to different ways the two-phase fluid is in contact with the active surface area of the sensor.
In this case, a notable drawback existscondensation is formed. Different methods exist which can be used to prevent the phenomenon. However, they are not perfect. On the one hand, they are not effective enough, and on the other, more importantly, they reduce the efficiency of the interaction of fog with a solid body. For our purposes, it is necessary to amplify the interaction to the maximum. When condensation appears, the layer of droplets will work against the registration of the following changes in the fluid composition. The surface will also get contaminated with time, which will interfere with the measured results. However, we should keep in mind that the layer may not prevent the registering of those composition changes if it absorbs them. This means that the layer will absorb and change, and thus the signal will also change. We will have to monitor the changes from a new ground level, which is inconvenient. The surface can be kept in satisfying conditions if used in a clean environment and if it isn't very sensitive. There is also the danger of different layers forming due to surrounding (atmospheric) conditions, the density of the fog in this spot and other unpredicted factors. Under different conditions, different signals will be generated, which will spoil the information we try to attain. It can be immediately noted that this layer will evaporate, contaminate and change with time, which is strongly undesirable.
We have a similar and already developed method for solid bodies and we have used it, for example, to detect counterfeit coins. The counterfeit coins have a different composition than the originals. Up to the moment, we do not have results clear enough to confirm the possibility of applying the method on fogs. There is also a problem concerning the amount of examined fog. To achieve correct results, the amount has to be identical to high precision for each measurement. This introduces serious technical difficulties. We have ideas for the development of sensors that would help ensure the desired volume, but the system will get too complicated. In fact, even if we can spray identical amounts of the fog into the examined volume, the fog will always precipitate to some extent, which will lead to errors. Difficulties also arise for moving volumes of fog from one place to another.
By using a filter, most of the impurities in the twophase fluid will be sorted out. As a result, its chemical composition will change. We can control the extent of this composition change of the filter via our developed method for solid bodies. The impediment, in this case, is that the filter will also hold a fraction of the liquid forming the fog. This will result in considerable errors in the results.
We have been experimenting and looking for a surface or a layer that is not sensitive to water and yet is sensitive to the additives. Strong candidates are

Filtering the fog
superhydrophobic coatings, including graphene, since it has strong hydrophobic properties. However, the information we have gathered and the results collected from conducted experiments did not encourage us to continue our research in this direction.
We have been trying to eliminate condensation by heating the examined structure. To a great extent, we were successful. However, the problem lies not only within condensation. The fog interacts with the work surface in an uneven manner, which yields an uneven signal. The density of the fog changes constantly. Depending on the amount of fog that has interacted with the structure, the signal will be different. We cannot be sure if the change in signal is due to the volume of the mixture or due to the change in density. We needed something that would eliminate the affection of the basic fog matter so that only the affection of formed contaminants remains.

Design possibilities for sensors for control of fog contaminations
After intense experiments and analysis of the gathered results, we concluded that the most convenient of the above-described methods is the third one. However, the function of the filter is fulfilled by a layer of liquid, which has an identical composition to the liquid forming the fog. If the filter is the liquid that the fog is comprised of, this liquid will not change when clean fog is sucked, but it will change with the introduction of a new agent. If a few drops of the filtered liquid are added, the signal does not change. An unwanted deviation in the signal may be observed if a considerable amount of the liquid is added. It is possible that the effect of the level sensor is present -dependence of the EMEE signal from the level of the fluid. This problem can be avoided by the creation of a structure that is insensitive enough to this effect or by automated replacement of the liquid before each measurement. Something else to consider is that when we feed fog towards the liquid, we won't be sure again of the exact amount. This is of no concern because clean fog does not change the signal. The signal depends only on the impurities. This allows for better accuracy.
We will review some possibilities for sensors for control of fog contaminations, which we have tested. In fact, besides the option of using a liquid as a filter, we have experimented with a large number of other structures providing the interaction between a twophase fluid and a solid. Our work in this direction continues. The most remarkable designs up to this moment are given in the following points.
The essence of this idea is that during our work we discovered the possibility for measurement of the EMEE-generated signal in a setup where the electrode, from which the signal is taken, is distant from the irradiated body. This is illustrated in Fig. 2.
A system for remote measurement of the EMEE signal The solid body (S) is exposed to modulated light. This generates an alternating electrical signal, which is captured by the electrode (E). Under suitable conditions, the distance to the irradiated body (S) can be more than several centimeters. The signal is registered by a measuring device (V). The fog we want to control is inserted in the area between the electrode (E) and the solid body (S). The body can be placed beneath a dielectric dome (D), which protects it from contamination and condensation. We have experimentally discovered that the signal generated by the EMEE is affected by the presence of two-phase fluid in the area between the body and the electrode. It is also sensitive to changes in fog parameters.
We have currently ceased work on this structure. At the current moment, it is unclear how to always supply identical volumes of fog with constant density. This will be a necessity for correct measurements. The questions of how to avoid condensation on the electrode and the protective dome and whether this condensed matter will have an influence on the results (along with other technical problems) have led to the fact that currently, we do not have clear and repeatable results on whether the received signal is strongly dependent on the composition of the fog.
The principle of this idea is derived from yet another possibility for EMEE signal registration we discovered. We are using an elongated specimen (S). One part of the specimen is concealed inside a metal container. The idea is illustrated in Fig. 3.
A device with differentiated areas of illumination and fog interaction On the metal container, there is a hole, through which the solid body (S) is inserted. The upper cover of the structure is made out of glass covered by a  transparent conductive layer for protection from electromagnetic disturbances. The part of the body inside the container is illuminated through the transparent cover. The signal is taken from an electrode (E), also inside the box. The interaction fog solid body happens outside of the box. Fog is being fed to the outside part of the body. Therein the interface (I) lies, generating changes to the signal proportional to changes in the fog. The difference in this setup is that the illumination area (L) and the body-to-fog interaction area (I) are more than 2-3 cm apart. We deduce that this distance can be much greater. This brings the possibility to isolate the irradiated area of the specimen from the effects of the fog (condensation, contamination, etc.).

A device with differentiated areas of illumination and interaction with the fog
With the described structure, we were able to register a well-defined influence of the two-phase fluid over the generated EMEE signal. Furthermore, the signal was also changing as a function of several changes in the fogcomposition, the diameter of the droplets, feed angle and so on. Those facts confirm that the device is suitable for sensor development. For the moment, we haven't implemented it for the following reasons. The problems of condensing and contamination of the work interface have to be solved. It is possible to create an automated surface cleaning. This will, however, make the system more complex. The illuminated specimen could be changed as a consumable, but our research showed that there are possible complications arising from surface inequalities. This problem will be regarded further on. Such inequalities can lead to large deviations in the measured signal. Further long-lasting tests are needed in this area. This concept will be valuable for us when it comes to sensors for two-phase fluids. It can help us to go around the difficulty of avoiding the influence of the fog over the intensity of the light beam illuminating the work surface.
The concept is shown in Fig. 4.
A concept project for a device with changing work surface At any given moment, only one of the plates is active (W). Only the active work plate is exposed to an electromagnetic field and has contact with the controlled fluid (M). The remaining workpieces are hidden inside a metal box and are being cleaned after measurement. Cleaning can be done in various ways. For example, mechanical cleaning via a sponge (D) or submerging in a container with a cleaning agent (C) along with other possibilities can be implemented. A drawback is that the cleaning element also has to be replaced periodically.
In the process of examinations, this concept has proven most preferable for the moment. It is shown graphically in Fig. 5. Over the substrate (S) there is a liquid layer (L). Again, this layer has the same composition as the liquid used for fog generation. The fog (F) does not change the composition of the liquid (L) when it precipitates over it. Consequently, there is no change in the amplitude of the signal generated by the interface (I) and sensed by the electrode (E). If contaminations are introduced in the fog, they will change the composition of the liquid (L). This will lead to changes in the interface (I) where the signal is formed. Thus, the signal will reflect those changes.
A structure with a liquid layer over the interface The essential fact here is that the contact surface of the sensor is the same as the basic compound of the fog. This way the signal is not distorted by the condensate. The liquid (with the same composition as the fog) eliminates the dependence on the amount of fog used (the basic fog matter does not change the signal). All that is left is the relation to the amount of impurities in the fog. The problem with the variable fog volume is avoided to a great extent.
The liquid layer over the semi-conductor serves also for signal accumulation. The signal needs to accumulate for a period of time, to eliminate fluctuations in the fog. For example, if a slight stream of air occurs, the amount of fog interacting with the active surface changesthis also changes the quantity of the controlled mixture and the signal deforms. Such processes are fastthey take place in a matter of seconds. By changing the quantity of fog, consequently, the quantity of admixture is also changed (since the admixture is contained within the fog). For this reason, a device is needed, which needs to average the signal over a predefined period. To be more exact, in our case we will measure the average quantity of admixture that has fallen over the sensor surface over a defined period. After that, the liquid is disposed of and automatically refreshed. This period  could be programmable. With such a device it was experimentally proven that an alteration in the EMEE signal is observed when the composition of the fog changes. It is important to note that once the parameters of the experimental setup are set, they need to be maintained constant. Only the variable under examination should be changed. In any other case, the possibility of errors is very high. An example of the overall reactance of such a device when fed fog mixed with 10% ammonia solution is shown in Fig. 6.
Reaction of a structure with a liquid layer over the interface during interaction with fog containing ammonia impurities We can see a distinct change in the amplitude of the signal generated by the EMEE. In this particular case, we have a decline in amplitude. This is defined by experiment conditions. There are many such specifics. This is why each structure to needs to be thoroughly tested under different parameters and design combinations. For example, we can create two devices, which have a specimen from the same semiconductor but are differently sized. If there is no contact with a liquid, the structure with the bigger specimen will generate a stronger signal. However, if exposed to water, the structure with the smaller specimen can create a signal many times stronger. Many factors contribute to such differences: the specific semi-conductive specimen, point of irradiation, overall design of the structure, irradiation intensity, etc.
As an example in this matter, we can review the relation to the laser beam intensity. We can introduce a new parameter: fog-solid interaction efficiency. It relates to the sensitivity of a given sensor structure to changes in the fog. It is dimensionless, expressed in percent -the ratio between the amplitudes of the EMEE signal before and after the change in the controlled parameter. It is calculated by the following formula: (1) where x is the input, y is the output and D is the deviation in percent. Fig. 7 illustrates the dependence of the interaction efficiency of the laser beam intensity in the range of 150 µW -18 mW. A semiconductor is used for a solid. The modulation frequency of the beam is 30.1 kHz and the wavelength -λ=650 nm. The fog is always fed in five sequential sprays from a manual atomizer which is fixed in a defined location. The liquid for fog generation is created when 0.2 ml of 5% iodine solution (potassium iodine, ethanol 96% and purified water) is dissolved in 20 ml of water.
Dependence between the fog-solid interaction efficiency and laser beam intensity It is obvious that for higher intensities of the beam there is a greater sensitivity. This has to be accounted for when real measurements are taken. But again, we need to note that this fact has to be investigated under many other combinations of parameters because it may change under other conditions. Dependence between the fog-solid interaction efficiency and laser beam modulation frequency Fig. 8 illustrates the abovementioned relation, but this time as a function of the laser beam modulation frequency. It is apparent that the modulation frequency is yet another contributing parameter and needs to be selected for each specific situation. A beam with λ=650 nm is used. The liquid substance for the fog is prepared by dissolving 0.2 ml of 5% iodine solution in 20 ml of water.
We have also confirmed that the fog-solid interaction efficiency varies greatly depending on the location of the laser illumination spot on the specimen. This called for continuous and laborintensive examinations of a large number of semiconductive plates. Research in this matter continues. A method is being developed for the express evaluation of the surface topography of the examined surfaces and the location of points of interest.
The examined structures have high sensitivity. If we introduce pure water (without inhibitor) and leave the system at rest, we will observe that the signal generated from the EMEE changes over time. This is so because the water itself is a live, functioning system with internal processes. The signal we monitor reacts to those processes. If the structure has no water,    there is no change in the signal. There are also no changes if we put ethanol instead of water because there are no ongoing processes in the ethanol. Those results indicate not only high sensitivity but also suggest that in this manner chemical and biological compositions can be monitored and controlled (including such used as weapons).
An experimental setup has been developed that is sufficient for conducting experiments regarding the development of sensors for the control of fog decontamination. A large number of variations of different sensor structures have been tested. The obtained results have been analyzed and a decision has been made based on them on the most suitable structure type that will be used for further research. The most notable structures that won't be currently in use have been also described. An overview has been made of their advantages and disadvantages.
It has been experimentally proven that a change in EMEE signal is observed when a change in the composition of the fog is present. It has been demonstrated that the signal generated by the sensor devices is influenced by a large number of conditions. When those conditions are combined properly, high sensitivity can be achieved. Those are means of improvement for the devices. For this purpose, thorough investigations can be made for each developed structure. Examples have been given on the relation between the fog-solid interaction efficiency and laser beam intensity, modulation frequency, etc.
Research results indicate the possibility to use these methods for the detection of chemical and biological agents, including such used as weapons. We hope that we will be able to use this sensor to perform control over all kinds of CBRN agents. After multiple searches, we weren't able to discover a similar device in existence worldwide.