Skip to main content

Improvement of construction materials used in the food industry to lengthening processing time (MODSTEEL)

Deliverables

For a wide range of commercial available surface modifications on stainless steel substrate with two different finishing (2R and 2B) the surface roughness was measured and represented by different surface texture parameters. Beside the usual surface texture parameters like roughness average Ra, root mean square roughness Rq, mean roughness depth Rz and maximum roughness depth Rmax, also values of the height of the highest profile peak Rp and the depth of the deepest profile valley Rv were measured. In addition to the pure roughness parameters some surface texture parameters that represent the material ratios of the roughness core profile depth Rk, the reduced peak height Rpk, the reduced valley depth Rvk and the smallest and highest material ratios Mr1 and Mr2 were tabulated.
Topography measurements on unmodified surfaces have been realised by UGINE&ALZ using Interferometric and Atomic Force Microscopes. For 2R finish (bright annealing), the average roughness Ra is of the order of 0.04µm and the total roughness Rt is of the order of 1-1.5µm.As for 2B finish, Ra is of the order of 0.1µm and Rt is around 2µm. Moreover, using AFM allowed to determine the approximate grain boundaries depth (about 0.5µm) and width (about 1 µm), as well as the grain roughness Ra (roughness without taking into account the grain boundaries, which is roughly about 50nm), of an annealed-pickled (2B) surface. Wettability measurements made by UGINE&ALZ show that the modified surfaces can be classified, in term of contact angle values, in the order: SiOx < implantation of SiF+ ions < DLC < Ni-P-PTFE. As far the surface energy is concerned, the order is the following: SiOx > implantation of SiF+ ions > DLC > Ni-P-PTFE. XPS results highlight the fact that the passive film on the 2B finish is thicker and contains more chromium than the one on the 2R finish, whereas such finishes have a similar contaminated film thickness. As for the alkaline detergent cleaning, it leads to a decrease in both the iron / chromium ratio and the passive film and contaminated film thickness.
Measuring the water contact angle tested different surface pre-cleanings and the cleaned surfaces used in the protein adsorption measurements. A cleaning protocol was defined and applied on the modified surfaces. The protein adsorbed amount obtained onto these surfaces was then studied.
Two Bacillus strains were used to test adhesion and ease of removal with the new surfaces. Data are available for each material and a comparison was done. The soiling was obtained in realistic conditions (milk at 60°C) and cleaning in industrial CIP conditions (NaOH 0.5% at 60°C).
Once optimum modified steel surfaces were prepared based on laboratory scale studies, initially in the form of a plate heat exchanger, its fouling and cleaning characteristics were tested by INRA on a one fifth of the actual factory scale. The most promising surfaces based on the fouling reduction and ease of cleaning were selected. Required adjustments were proposed based on how the coatings proposed withstand fouling conditions when considering temperatures above 100°C, and cleaning in place conditions using NaOH at 0.5% V/V.
Transfer tests performed by UGINE&ALZ show that the Silica and Excalibur coatings release little Si in soda. As for Ni-P-PTFE coating, it releases Ni in acetic acid, it is almost completely dissolved in nitric acid and it also releases Ni in the dairy products, which can cause dermatitis to those people ingesting such products. This Ni release of Ni-P-PTFE coating in dairy products has been checked several times, which means during INRA (V2 and Clip 6 scales) and DANONE fouling / cleaning tests. The Ni-free hydrophobic coating made by UGINE&ALZ has a different behaviour. There is no release in the dairy product; little release in soda only occurs. Formability tests have also been implemented by UGINE&ALZ. Ni-P-PTFE, Xylan and the Ni-free hydrophobic coating present many cracks after such a test. Yet, no removal of such coatings due to the deformation seems to occur, which means that the adherence between the substrate and the coatings is rather good.
Impact of surface modifications on the kinetics of fouling (reduction of the hydraulic diameter and of the global heat transfer coefficient). Consequences on the possible increase of processing time for sterilisation of desserts in plates heat exchanger.
This part of the project analysed the relationships between the basic component of the surface tension (gamma) of stainless steel and: - The build up of calcium phosphate deposits in the presence and in the absence of citrates; - The initial adhesion of whey proteins, - The build up of calcium phosphate/protein deposits. The different stainless steel surfaces were characterized by the sessile drop method (measurements of contact angles formed by three appropriate liquids on the surfaces). It was found that the basic part of the Lewis acid-base component (gamma) could be related to the deposition features of calcium phosphate under several morphological phases, with and without whey proteins. The specific conclusions related to the influence of the surface on the deposition process were: - SMUF deposition without whey protein: -- The initial deposition rate of calcium phosphate as well as its easiness of removal decrease as gamma increases (the highest gamma corresponds to SiOx PECVD and ion implanted surfaces; the lower value of this component corresponds to the Ni-PTFE and silica coated surfaces); -- The final amount and density of the deposit increases with the increase in gamma; - SMUF deposition with whey protein and whey proteins alone: -- For higher temperatures, the amount of deposit formed decreases as gamma increases (the highest gamma corresponds to SiOx PECVD and ion implanted surfaces; the lower value of this component corresponds to the Ni-PTFE and silica coated surfaces). -- For deposition at lower temperatures, the dependence on the surfaces is similar to the one found for the deposition of calcium phosphate alone. - SMUF deposition without citrate: -- The amount of deposit formed increased with the increase of gamma- (the highest gamma corresponds to SiOx PECVD and ion implanted surfaces; the lower value of this component corresponds to the Ni-PTFE and silica coated surfaces).
The effect of the surface on the deposition behaviour of calcium phosphate was quantified in the presence and absence of protein as well as in the presence and absence of citrates. It was concluded that surfaces having lower values of Gamma- promote the formation of less and bigger aggregates in the initial phase of deposit formation. Opposed to surfaces having higher values of Gamma- where more although much smaller aggregates are formed.
The surface properties of a dispersion coating depend strongly on the distribution of the incorporated particles. For instance, in the case of the auto-catalytic Ni-P-PTFE coating, the fouling behaviour of the surface improves only with the quantity of PTFE in the surface, if the PTFE-particles are homogenously distributed. To develop new coating systems and to investigate their properties it is necessary to have a possibility to detect the location of the particles in the surface. With an enhanced AFM technology a way was found to obtain information about the particle distribution in the outer surface of a Ni-P-PTFE coating. The same technique may also be used for almost all kinds of dispersion coating systems. The advantage as compared to REM measurements is the possibility to obtain with the AFM two completely independent images. One represents the topography and the other the particle distribution. This technology will even be applicable down to micro-scale. In the present investigation PTFE-particles with a diameter of 0.2µm have been detected without any problems.
Simulated Milk Ultra-Filtrate (SMUF) solutions were employed to investigate calcium phosphate precipitation characteristics, under various conditions, in two types of tests. Spontaneous precipitation of calcium phosphate was studied in narrow ranges of pH (5.7 to 7.0) and temperature (55 to 75 Co). The morphology and size of precipitated particles depends primarily on solution super saturation; reduction of pH leads to relatively larger particles. The precipitated solids are identified as poorly crystallized hydroxyapatite. The presence of citrate ions results in morphological changes of precipitated crystals. Calcium phosphate precipitation was also investigated under continuous flow conditions by employing a tubular test section, at temperatures 60 and 70 Co, pH 5.8 to 7.0 and two Reynolds numbers (7.000 and 17.000). Deposition rate is affected by both temperature and pH, through solution super saturation. At the maximum deposition rate, an initially thin layer is formed, covering entirely the substrate within minutes from the onset of flow. Further deposition occurs mostly in the form of overgrowths consisting of clusters of sub micron particles. With increasing flow velocity the deposition rate tends to increase, and more compact deposit layers are obtained. These tests did not provide clear evidence that the deposited mass is affected by the surface treatment or type of substrate; however, SEM deposit observations indicate that adhesion strength of modified surfaces might by reduced, compared to steel substrate, possibly facilitating clean up. The techniques employed in this work are useful in testing heat exchanger surface characteristics as regards deposition in dairy and other processes.
Coating of commercial stainless steel surfaces (finish 2R and 2B), employed in the manufacture of heat exchangers, has been achieved by ultra thin (sub-micron thickness) layers, using a sol-gel technique. Defect-free silica coatings have been obtained with two notable characteristics: - hardness greater than that of the stainless steel substrate. - reduced surface roughness compared to the (bare) substrate. Titania films were also obtained by the same techniques. A promising procedure for obtaining mixed particulate/sol-gel titania coatings were developed in order to cope with film cracking. Although film cracking was not completely eliminated, employing stable concentrated dispersions of titania nanoparticles drastically reduced it. The knowledge gained in the preparation of these nanoparticles, sols and coatings is considered useful in future endeavors of producing particles and modified surfaces with improved properties. Using equipment developed at CPERI, coatings were produced on medium size surfaces and heat exchanger plates (e.g. 15x35cm) as well as in tubes 50cm in length. Coating of larger surfaces and of longer tubes does not appear to be a difficult task. The coating technique, demonstrated at laboratory scale, is relatively inexpensive and reliable.
Tests carried out with biofilm and SMUF deposits showed that the biological layer masked the differences between the different surfaces but, at the same time, increased the amount of total deposit formed. Also, no significant differences were found between the number of total cells (DAPI) and active cells (CTC and Live/Dead) attached to the different surfaces.
The influence of the surface properties and the bulk solution conditions on the adsorption behaviour of whey proteins onto the steel surfaces was investigated. Further, the protein aggregate size and distribution present at the different bulk conditions was determined and related to its adsorption at the surface.
Assessment of the possibility of adapting the cleaning procedures towards the modifications of the surface properties of plates (MOS2 bombardment and NiPPTFE coating) for sterilisation plates heat exchanger (dessert applications).
The rotating disk apparatus showed to be a very useful tool for predicting fouling results in a faster and easier way, because the results obtained in the heat flow cell follow the same trend as those obtained for this equipment.