As mentioned above, microbial adhesion is a complex biological process that is influenced by the physiological features of the microorganisms, the environmental properties and the physicochemical properties of the surface (Yi et al., 2004; Moons and Michiels, 2009; Whitehead and Verran, 2007; Götz, 2002). The results of the studies indicate (Monds and O'Toole, 2009), that the nutrients for the microorganisms of the biofilm are contained in the solution, so the composition of the nutrient medium, its ionic energy, pH, temperature also affect the adhesion intensity of the microorganisms to the base. The formed biofilm consists of an inhomogeneous structure, resulting in a gas concentration gradient, in particular, a decrease in the amount of oxygen from the periphery to the depth, pH gradients and nutrients. These gradients of biofilm functioning provide physiological variability between individual biofilm cells - they grow more slowly at the depth of the cell than at the periphery of the film, leading to phenotypic stability and a dramatic change in environmental factors. The formation of the biofilm and the strength of its adhesion to the surface are influenced by the hydrophobic properties of the surface of the microorganisms, the presence of their villi and flagella. Most bacteria have a negatively charged surface that contains hydrophobic external components that cause hydrophobic interaction with the substrate. The presence on the surface of biofilms of polysaccharides or proteins helps bacteria to adhere to the surface, thereby providing a competitive advantage in the formation of biofilms for specific cells of the microbial association (Yi et al., 2004). In addition, the structure of the biofilm is influenced by the presence in its composition of microorganisms, which in the course of their growth and development release the gaseous substances of metabolism. This leads to a decrease in the density of the biofilm and the uneven flow of nutrient substrates (Götz, 2002). Among the many factors that influence the adhesion process, researchers highlight the role of surface properties, which is defined as the most significant (Zogaj et al., 2003; Kania et al., 2007). As a result, three theories of microbial adhesion to the surface were proposed: thermodynamic, DLVO theory and XDLVO (Kolari, 2003; Langsrud et al., 2016). The thermodynamic theory is based on the fact that when the microorganisms are attached to the surface there is a change in the total free energy of Gibbs, Van der Waals forces. Theory DLVO is based on the fact that the colloidal particles of the lyophobic dispersion system can move closer to each other without interruption until their liquid diffuse shells come into contact. Theory XDLVO is based on thermodynamic and DLVO theory (Langsrud et al., 2016). However, researchers believe that all three theoretical models designed to reveal the essence of microbial adhesion to the surface have been designed for the perfect colloidal system. Under production conditions, microbial adhesion is a much more complicated process and attachment of microorganisms can occur in different ways.

In the dairy industry stainless steel corrosion-resistant steels of the following brands are most often used for the equipment AISI-304, AISI-316, AISI-321 (Jullien et al., 2003; Dantas et al., 2016). These steel delivery conditions can have different surface roughness from 0.2 – 3.2 mm. Studies to investigate the effect of terrain and surface roughness on microbial adhesion are not straightforward. According to research (Dantas et al., 2016), there is a correlation between surface roughness and bacterial adhesion, the attachment of microorganisms to the surface increases with increasing roughness. From the analysis of works (Kukhtyn et al., 2016) found that the formation of biofilm was much slower on the surface with a roughness of up to 0.4 μm, compared with a roughness greater than 0.8 μm. Using electron microscopy, the researchers (Dantas et al., 2016; Kukhtyn et al., 2016) found that the primary adhesion of microbial cells occurs along the hollows of the surface roughness, since under these conditions the area of contact of the microbial cell with the surface increases. However, other studies indicate that there is virtually no correlation between stainless steel surface roughness and microbial adhesion. There are also various data on the effect of surface wettability on microbial adhesion. It was found that the number of adherent bacteria decreased with increasing surface hydrophobicity, and microorganisms that attached to hydrophobic surfaces were more easily removed by increasing the flow force during fluid circulation (Dou et al., 2015). However, other researchers (Kolari, 2003), indicate that there is no correlation between surface wettability and microbial adhesion. It is investigated that formed biofilms on surfaces with high roughness reduce the efficiency of heat transfer in condenser heat exchangers by about 15%. Thus, the hygienic quality and cleanliness of the processing equipment in the dairy industry after sanitation can be closely linked to relief and roughness. Therefore, research into the process of film formation on stainless steel with different roughness over time and using different shapes and sizes of bacteria is promising. These studies will allow us to understand more deeply the process of the formation of microflora on the equipment and, accordingly, the contamination of food products. Also, finding out the effect of surface roughness on microbial adhesion will help to improve and modify the surfaces that interfere with adhesion.

The purpose of the work was to determine the effect of different surface roughness of stainless steel on the process of microbial adhesion and film formation, depending on the physiological and morphological features of the microorganisms that contaminate the equipment.

To achieve this goal, the following tasks were set:

– to experimentally investigate the process of film formation by rod-shaped and coccoidal bacteria in stainless steel with different surface roughness at a temperature of 25 °C;

The successful formation of microbe biofilms is not possible without microorganisms’ surface adhesion. The scientists have separated five main stages of biofilms formation and development on any surface. Bacteria surface adhesion depends on the initial number of microorganisms, their shape, temperature, environment pH, size and etc. The primary adhesion of microbial cells is taking place along the cavities of the surface roughness as under such conditions the contact area of the microbial cell and the surface is getting bigger. Nevertheless, any process of each biofilm formation has specific features that are regulated by both genetic, biochemical properties of the bacterial cell and environmental factors as well. We may claim that a biofilm is a microbial combination formed by the cells attached to the surface and to one another and are in the matrix of synthesized extracellular substances. The most serious biofilm hazard on the milk processing equipment is that the biofilm extracellular matrix protects the bacteria from the disinfection agents’ actions. The survived microorganisms occupy the new surfaces and milk products. Thus, while studying the microbial adhesion and biofilms formation under laboratory conditions one can not always take into consideration the impact of production factors and physiological specific features of microbial cells. Conducting the research on studying the process of film formation on the stainless steel of different surface roughness during a certain period of time and using different bacteria forms and size will allow us to understand deeper the process of microbial flora formation on the equipment and also food contamination. To find out the impact of surface roughness on microbial adhesion means to facilitate the improvement and modification of the surfaces which resist the adhesion.

methods Experimental studies were carried out using modern standard and conventional methods: technical (determination of surface roughness of stainless steel), microscopic (light and electron microscopy of the process of film formation and degradation), spectrophotometric (optical density), biofilm, microbiology, microbiological parameters of dairy equipment and dairy products. Investigation of the process of film formation by microorganisms on the surfaces of metals with different roughness was carried out in a sterile Petri dish. The sterile stainless steel plates with appropriate surface roughness were placed in the cup and sterile meat-peptone broth was introduced into the cupand the appropriate test culture (Staphylococcus aureus, Escherichia coli, Enterococcus faecalis or any other dedicated dairy equipment) in concentration to 1 cm² plate area accounted for from 1 to 50 thousand cells. The incubation was carried out at temperatures from +17 °C, +25 °C to +37 °C for 3, 6, 9, 12, 18 and 24 hours, depending on the purpose of the experiment. After incubation, the plates were removed from the Petri dishes, washed three times with planktonic (unattached) microorganisms with phosphate buffer. Recorded the formed microbial biofilms on the plates with 96% ethyl alcohol. After fixing, the biofilms were stained, for this purpose the plates were immersed in 0.1% crystalline violet aqueous solution. After staining, the plates were washed three times with phosphate buffer and dried. After drying, each plate was individually poured into 7.0 cm³ of 96% ethyl alcohol and left for 20 min. After 20 min exposure, 5 cm³ of the wash solution was removed from biofilms and the optical density t 570 nm was determined spectrophotometrically on the KFK-3 photometer (Ukraine) (Figure 1).

Electron microscopic studies of the process of forming biofilms on stainless steel were performed on an electron raster microscope (RЕМ 106 I, Ukraine) with magnification from 2000 to 5000 times (Figure 2). In addition, the adhesion of microorganisms in the cavities or roughness projections was visually assessed using a microinterferometer МII-4Y4.2 with magnification in 1500 times (Figure 3).

To determine the parameters of the surface layer, the following technique was used:

- scanning the surface with a 3D scanner David Laser Scanner SLS-1 using the software David Laser Scanner Professional Edition;

A large number of dots were merged into the surface when scanned. To determine the roughness and waviness of the surface layer, we cut out a scanned area of 10 x 10 mm in software PowerShape, which is shown in Figure 4. We draw two two perpendicular planes (Figure 5) that will cut the plot. A fragment of a longitudinal profile with points (vertices and cavities) is shown in Figure 6. Similar operations were performed with a transverse profile, a fragment of which with vertices and cavities is shown in Figure 7. Determining the lengths of perpendiculars from cavities or vertices to the center line using the formula:

R a = ∑ i n | Y i | n ,

Where:

n – the number of points (in our case 107); ∑ i n | Y i | – the sum of the lengths of all perpendiculars (in our case 2038 μm).

So,

R a 1 = 203.8 107 =1.905 ( microns ). R a 2 = 104.7 64 =1.6 ( microns ).

Accordingly, the arithmetic mean, is:

R a = R a 1 + R a 2 2 = 1.905+1.635 2 =1.77 ( microns ).

Using the standard table of accepted values of Ra, we can determine that the test area has a roughness within Ra 1.6 and Ra 2.0.

To obtain mathematical dependencies when comparing the results of the study of optical density by the factor of time and roughness for different temperatures and the initial bacterial count, regression equations of optical density and bacterial development were proposed.

Optical density regression equation (quantity) Y by time factors and roughness r for different temperature ranges T °С and the initial bacterial count according to Table 2 will be:

Y k,i  =  y 0  +  y 1 ⋅ r k + y 2 ⋅ t 1 + y 3 ⋅ r k ⋅ t i + y 4 ⋅ ( r k ) 2 + y 5 ⋅ ( t i ) 2

The regression equation of the rate of change of quantity Z by time factors tand roughness r for different temperature ranges T °С and the initial bacterial count according to Table 3:

Z k,i  :=  z 0  +  z 1 ⋅ r k + z 2 ⋅ t i + z 3 ⋅ r k ⋅ t i + z 4 ⋅ ( r k ) 2 + z 5 ⋅ ( t i ) 2

By comparing population intensities E. coli for Т = 37*I(n<1), I(2<n<10), I(20<n<50) got that Y(2<n<10)_k,I = s1_k,i* Y(n<1)_k,I, Y(20<n<50)_k,I = s2_k,i* Y(2<n<10)_k,I,

Where: s 1 = ( 1 1 1 1 1.056 1.139 1.152 1.252 1.167 2.075 1.777 1.346 1.201 1.435 1.455 1.554 1.698 1.776 1.929 1.252 1.625 1.472 1.298 1.087 1.278 1.112 1.104 1.09 ) s 2 = ( 1 1 1 1 1.053 1.165 1.496 1.396 1.889 1.198 1.514 1.56 1.6 1.178 1.354 1.226 1.138 1.085 1.066 1.363 1.051 1.125 1.027 1.199 1.05 1.089 1.02 1.133 )

Where:

k = 0 – 6; i = 0 – 3.

Thus, the initial number of bacteria is taken into account as a factor.

Optical density regression equation (quantity) Y by time factorstand roughness r for different temperature ranges T °С and the initial bacterial count according to the Table 4:

Y k,i  =  y 0  +  y 1 ⋅ r k + y 2 ⋅ t 1 + y 3 ⋅ r k ⋅ t i + y 4 ⋅ ( r k ) 2 + y 5 ⋅ ( t i ) 2

The regression equation of the rate of change of quantity Z by time factorstand roughness r for different temperature ranges T °С and the initial bacterial count according to the Table 5.

Z k,i  :=  z 0  +  z 1 ⋅ r k + z 2 ⋅ t i + z 3 ⋅ r k ⋅ t i + z 4 ⋅ ( r k ) 2 + z 5 ⋅ ( t i ) 2

By comparing population intensities E. faecalis for Т = 37*I(n<1), I(2<n<10), I(20<n<50) got that Y(2<n<10)_k,I = s1_k,i* Y(n<1)_k,I, Y(20<n<50)_k,I = s2_k,i* Y(2<n<10)_k,I,

Where:

s 1 =( 1 1 1 1 1.056 1.111 1.214 1.311 1.229 1.759 1.686 1.57 1.186 1.279 1.202 1.405 1.899 1.794 1.854 1.338 1.585 1.486 1.3 1.077 1.293 1.144 1.062 1.09 ) s 2 =( 1 1 1 1 1.042 1.133 1.441 1.729 2.007 1.459 1.49 1.404 2.039 1.259 1.266 1.361 1.118 1.084 1.136 1.307 1.104 1.129 1.036 1.203 1.051 1.061 1.043 1.128 )

According to the regression equation, we construct response surfaces (Figure 10, 11 and 12).

Figure 10 (Y) shows that the number of bacteria Е. coli highest value gets at roughness surface 2.68 μm after 20 hours of incubation. The speed of development (Z) Е. coli increases most intensively with roughness values between 0.95 μm and 2.68 μm and time, starting after ten hours of incubation. Analysis of Figure 11 shows that similar trends in bacterial counts and growth rates, as well as in temperature +17 ±1 °С. However, the number of bacteria is increasing (Y) and fills roughness (Z) from 2.68 microns to 0.63 microns at 18 hour of incubation, and at 20 hour is practically up to 0.30 microns. From the temperature data +37 ±1 °С the number of bacteria is growing rapidly (Y) and the surface roughness is filled (Z) 2.68 – 0.63 microns within 12 hours of incubation. On the basis of mathematical modeling, it was found that the adhesion and intensive process of biofilm formation Е. coli passes in the hollows of great roughness 2.68 – 0.95 micronsand it gradually fills the hollows with less roughness 0.63 – 0.16 microns. Figure 12 shows that there is a similar pattern in the process of film formation for E. faecalis, as in Е. coli at this temperature. However, the rate of growth E. faecalis іs more intense as the roughness depressions fill from 2.68 microns to 0.63 microns as opposed to Е. coli to 0.95 microns. The process of film formation in E. faecalis for temperatures +37 ±1 оС, on the surface with different roughness, was more intense compared to Е. coli, that is, the rate of filling all the roughness depressions took 12 hours. The data obtained indicate that the biofilm density depends on the initial number of microbial cells on the steel surface. It has been shown that the more contaminated the surface with microorganisms, the faster the process of film formation and formation of dense biofilms.