Anti-Fouling Behaviors of a Modified Surface Induced by an Ultrasonic Surface Rolling Process for 304 Stainless Steel

Abstract

This research aimed to investigate the effects of an ultrasonic surface rolling process (USRP) on the deposition behaviors in CaCO₃ crystallization fouling for 304 stainless steel (304SS). The microstructure, surface morphology, and hydrophobic properties of the modified layer were characterized with optical microscopy, a roughness profile measurement instrument, and a contact angle measurement instrument. The corrosion and fouling behaviors of different surfaces were studied in simulated cooling water. The polarization curves and electrochemical impedance spectroscopy (EIS) were used to evaluate the electrochemical properties. The results showed that USRP-treated surfaces had better anti-corrosion and anti-fouling performance. The improvement in anti-fouling performance was attributed to the weakening of peaks and valleys, the reduction of surface defects, and the improvement of corrosion resistance.

1. Introduction

Heat exchangers are widely used in industries, such as the petrochemical, chemical, power, automotive, and aerospace industries, as well as in seawater applications, food processing, and medicine. Fouling is widely present in heat exchangers. Fouling induces a heat transfer loss and substantial pressure drops and under-deposit corrosion, resulting in reduced performance. Fouling is a very important reason for the failure of heat exchanger tubes [1]. Crystallization fouling is the most common type in heat exchangers because water that contains mineral salts (CaCO₃, CaSO₄, etc.) is the most common medium used in such processes. The proportion of CaCO₃ is more than 80% on the scale. Among all types of fouling, crystallization fouling is the most harmful to heat exchangers [2]. The economic losses for fouling prevention and removal every year are notable, and environmental damages are also incurred.

Several approaches have been adopted to reduce fouling deposition, including chemical, physical, and mechanical methods. Among them, surface modification is identified as the most promising technology [3], and low cost is also an important reason for its attraction. At present, two methods are commonly used to modify surfaces—chemical coating and physical modification, which is mainly through mechanical processing. There are some problems with the thermal stability, wear resistance, mechanical strength, and durability of chemical coatings. Therefore, the emphasis in anti-fouling is a focus on developing stable and environmentally friendly surfaces. Compared with surfaces that undergo chemical methods, the surfaces modified with physical methods are environmentally friendly and display superior mechanical strength, thermal stability, and low thermal resistance. There are many methods for physical modification, such as electrical discharge machining [4], shot peening [5], supersonic fine particle bombarding [6], and the ultrasonic surface rolling process [7]. Recently, physical modification technologies were used in the medical field. Studies have shown that the performance of metal materials treated by severe shot peening can be modulated in terms of biocompatibility, adhesion of bacteria strains, and enzyme stability [5,8,9]. Physical modification technologies provide a new idea for anti-fouling. By using physical modification technologies, it is possible to solve the problem of fouling by using the characteristics of the substrate.

Crystal adhesion is constrained by the surface properties [10], such as the surface morphology, surface roughness, wettability, and corrosion resistance. Generally speaking, fouling is more likely to be deposited on rougher surfaces [2]. However, another study showed that the weight of fouling deposition was not linear with the surface roughness values [11]. The difference in the results may be due to the micro-textures and chemical compositions on surfaces. Compared with surface roughness, wettability has a more significant effect on fouling deposition [12]. Moreover, another study showed that anti-fouling performance had much to do with the corrosion resistance of materials [13]. The crystal type and the growth rate of fouling could be affected by different types of corrosion. The nucleation of CaCO₃ crystals can too easily occur on a surface if the material is prone to being corroded [14,15].

The crystallization and adhesion forces of CaCO₃ crystals can be modulated by changing the surface structure and composition of materials through physical modifications. The shot peening process can enhance the stress corrosion resistance, but it accelerates the propensity for fouling in terms of an increase in surface roughness and wettability [12]. A surface modified by an electrical discharge machining method had better anti-fouling performance than a polished surface, which was attributed to an increase in the corrosion resistance and hydrophobicity properties [4]. The USRP is a unique method of mechanical surface treatment that combines ultrasonic impact and static extrusion force [16,17]. It can provide superior surface mechanical properties and surface integrity by decreasing the surface roughness, refining grains, inducing residual compressive stress, and increasing hardness. Compared with other plastic deformation modification techniques, the USRP can also provide an appreciable reduction in surface roughness [18]. A recent study showed that the USRP was able to prevent the corrosion of an austenitic stainless steel surface due to a stable passive film and its small surface roughness [19]. The residual compressive stress induced by plastic deformation during the USRP played a beneficial role in the resistance to corrosion [20]. It was found that the hydrophobicity of a surface modified with the USRP was enhanced in our previous study.

In this research, 304 stainless steel was modified with the USRP. The microstructure, geometrical morphology, wettability, corrosion resistance, and anti-fouling performance of the surfaces were studied. The aim of the present study was to investigate the anti-fouling behaviors of a USRP-treated surface. The results can provide new ideas for the design and preparation of anti-fouling metals.

2. Materials and Methods

2.1. Materials and Reagents

A 304SS plate with a thickness of 2.5 mm was modified with the USRP on one surface. The chemical compositions of the substrate are listed in

Table 1. Chemical compositions of 304SS.

Element	C	Si	Mn	P	S	Cr	Ni
wt.%	0.04	0.42	1.15	0.03	0.001	18.22	8.02

. The static pressure, rolling line speed, and step size were 0.1 MPa, 2 m/min, and 0.05 mm, respectively. The modified metal was ultrasonically cleaned in ethanol and acetone for degreasing. For the experimental investigation, artificial hard water with CaCO₃ was prepared by dissolving a mixture of CaCl₂ and NaHCO₃ powders in distilled water, which was used to simulate closed recirculating cooling water. The concentrations of CaCl₂ and NaHCO₃ were 444 and 672 mg·L⁻¹, respectively, and the pH value of the solution was about 7.5–8.5.

2.2. Characterization

The surface morphology was measured via a sigma500 scanning electron microscope (SEM, ZEISS, Oberkochen, Germany) and a roughness profile measuring instrument (V1.0 2000, Shanxi Weier Mechanical and Electrical Technology Co., Ltd., Xian, China). The roughness parameters included the arithmetical mean of the deviation of the assessed profile (R_a), the maximum profile peak height (R_p), the maximum profile valley depth (R_v), the maximum height of the profile (R_z), and the skewness of the assessed profile (R_sk). The sampling length was 0.8 mm and the evaluation length was 4 mm for all samples. Water contact angle tests were performed with the sessile drop method while using a contact angle measurement instrument (JC2000, Shanghai Powereach Digital Technology Equipment Co., Ltd., Shanghai, China). All measurements were carried out at room temperature with a deionized water droplet of 2 μL, and the droplet remained on the surface for 10 s. The average value from five random regions was taken as the final test result. In order to observe the microstructure, transversal surfaces of the samples were first ground with sandpapers from a 180 mesh to a 1200 mesh, and then polished with polycrystalline diamond polishing agents. The metallographic structure was observed by means of optical microscopy (OM, ZEISS, Oberkochen, Germany). The samples were etched with aqua regia before the OM observations.

2.3. Electrochemical Measurement

The corrosion behavior was assessed with an electrochemical workstation (Interface 1000, Gamry, Philadelphia PA, USA). A three-electrode system was used in the tests, and the sample, a saturated calomel electrode, and a platinum electrode served as the working electrode, reference electrode, and counter electrode. During the electrochemical measurements, saturated KCl/AGAR was used as a salt bridge between the reference electrode and solution. These samples were sealed up by epoxy resin with an exposed surface size of 1 cm × 1 cm. After sealing, the untreated samples were polished to #1000 with sandpaper, while the USRP-treated samples did not need to be polished. Before each measurement, the samples were ultrasonically cleaned in ethanol and distilled water, and then dried. The electrochemical experiments were carried out in the artificial hard water at room temperature. The measurements were performed at the open-circuit potential. For the measurement of potentiodynamic polarization, the potential was scanned from −0.55 to 0.85 V for the untreated samples and from −0.41 to 0.98 V for the treated samples while using a scan rate of 1 mV·s⁻¹. The EIS tests were conducted over a frequency range from 10⁻² to 10⁵ Hz with a 10 mV signal amplitude. The impedance spectra were analyzed with the Zview 3.1 software, and the curves were drawn by using the Origin 2021 software.

2.4. Fouling Experiment

Before each measurement, the plate was cleaned by using ethanol and deionized water, and then it was dried and weighed with an analytical balance. The samples with a size of 30 mm × 30 mm × 2.2 mm were hung in the artificial hard water at 70 ± 0.5 ℃ and kept for 30 h. The samples were taken out and weighted every 5 h, and the fouling weight was calculated. The average value of three samples was taken as the final result for each material. The other groups of samples used to study the fouling morphology on the surfaces were immersed in the solution for just 20 min. The crystals on the surfaces were investigated with SEM (sigma 500, ZEISS, Oberkochen, Germany). SEM was able to provide a resolution of 0.8 nm. The SEM parameters are displayed at the bottom of the figures, including the magnification (Mag), acceleration voltage (EHT), and working distance (WD).

3. Results and Discussion

3.1. Surface Profile and Wettability

The roughness profile is shown in Figure 1. The roughness parameters of the surfaces are listed in

Table 2. Roughness parameters of the surfaces.

Specimen	Ra/μm	Rp/μm	Rv/μm	Rz/μm	Rsk
Untreated	0.190	0.466	1.108	1.574	−1.014
Treated	0.014	0.048	0.052	0.100	−0.166

. The value of R_a decreased from 0.190 μm for the base sample to 0.014 μm for the treated sample, and the values of R_z and R_v decreased significantly. The values of the parameter R_sk were strongly influenced by independent peaks or independent valleys. They were −0.166 and −1.014, respectively, for the treated and untreated surfaces. The value of R_sk of the USRP-treated surface was negative, and the absolute value was close to 0, indicating that the distribution of peaks and valleys on the surface was relatively uniform and the contour was smooth. The peaks and valleys of the treated surface were obviously blunter. Figure 2 shows the microscopic morphology of the test samples measured with SEM. For the original material, the surface was covered with grooves produced by the machining. Although there were still some defects, the surface of the USRP-treated sample became very smooth. Continuous plastic deformation played a predominant role in cutting peaks and filling valleys during the USRP. It is difficult for fouling to grow and deposit on smoother and flatter surfaces [21]. The USRP treatment reduced the nucleation points of crystal fouling by weakening the wave peak.

Pictures of the measurement of the contact angle are also shown in Figure 1. The contact angle increased from 72° to 100° after the USRP treatment. The nucleation of CaCO₃ was mostly heterogeneous nucleation on the heat transfer surface. According to the classical theory of homogeneous nucleation of crystals, the nucleation energy ΔG* on the plane substrate can be expressed by Equation (1) when the shape of a crystal’s nucleus is regarded as a spherical crown [22].

$$\Delta G^{*}=\frac{16{\mathsf{\pi }\mathsf{\Omega }}^2{\gamma }_{\mathrm{cv}}^3}{3\Delta g_{\mathrm{v}}^2}f(\theta )$$

(1)

$$f(\theta )=\frac{{(1-\cos \theta )}^2(2+\cos \theta )}{4},0^0<\theta <{180}^0$$

(2)

where Ω, Δg_v, and γ_cv are one atomic volume, the free-energy changes of a phase transition for an atom, and the interface energy between the crystal and fluid, respectively. θ is the contact angle. Theoretically, the larger the contact angle θ is, the greater the nucleation energy ΔG* is. Therefore, the nucleation rate of a hydrophobic surface should be lower than that of a hydrophilic surface, while showing a certain scale resistance. The contact angle of 304SS increased after the USRP treatment, which played a beneficial role in preventing fouling.

3.2. Microstructure

The metallographic images of the transversal sections with a magnification of 200 are shown in Figure 3. The microstructure of the matrix was a γ-austenitic structure, as shown in Figure 3a. For the treated sample, two zones were observed, namely, a deformation zone and a matrix, as shown in Figure 3b. There was no clear dividing line between the plastic deformation layer and the matrix. The thickness of the deformation layer was about 140 μm. There was a refinement layer of about 10 μm near the surface. In addition to many twins, deformation-induced αʹ-martensitic was observed in the USRP-treated surface. High-density dislocations were formed by severe plastic deformation during the USRP [8,23]. The grain refinement of 304SS was mainly achieved by a sliding dislocation network and twin segmentation that were formed by the dislocation transfer on the moving surface because of the low stacking fault energy [24].

3.3. Corrosion Performance

The polarization curves of the untreated and treated surfaces had the same trends, including a passivation zone and a pitting breaking potential, as shown in Figure 4. The corrosion potential (E_corr) of the treated sample was higher than that of the untreated one, which was increased from −157 to −88 mV. The current density (I_corr) obviously decreased under the action of the USRP. The smaller value of I_corr for the treated sample indicated a lower dissolution rate of the passivation film. The critical pitting potential E_pit, which was the potential at which the current density reached 100 μA·cm⁻², increased from 0.85 V for the untreated sample to 0.97 V for the treated one. The increased E_pit revealed that the pitting corrosion resistance of the USRP-treated sample was improved in hard water. Severe plastic deformation was able to result in a high surface residual compressive stress during the USRP treatment. The residual compressive stress was able to inhibit the metastable pitting in nucleation [20]. The residual compressive stress made the passive film more robust and compact because of the slightly lower interatomic distances in the underlying compressively stressed metal lattice [25]. This caused a lowered kinetics of the migration of point defects through the passive film, resulting in a lower current density value.

Nyquist and Bode phase plots for the untreated and treated surfaces are shown in Figure 5. The semicircular arcs of Nyquist diagrams are usually used to interpret electrochemical reaction processes. As shown in Figure 5a, a single capacitive arc was expressed for the untreated and treated samples. However, the radiuses of the capacitive arcs of the treated samples were much larger than those of the untreated surfaces, which indicated that the total resistance of the passive films was greater. Figure 5b shows that the impedance value and phase angle of the treated samples were increased compared with those of the untreated samples, and the corresponding phase angle peak was also widened, indicating that the stability of the passive film was improved. The equivalent electrical circuit (EEC) shown in Figure 6 was selected to fit the test data. In the EEC, R_s is the solution resistance, R_ct is the charge transfer resistance, and CPE is a constant phase element that was used to replace the capacitor element C. The capacitor element C is generally replaced by a CPE in an EEC due to the existence of the ‘‘dispersion effect” [26]. The CPE is composed of the double-layer capacitance C_d and the dispersion coefficient n. Then, the total electrode impedance Z can be expressed as follows [27]:

$$Z=R_s+\frac{R_{ct}}{1+{(2\pi f{R}_{ct}{C}_d)}^n}$$

(3)

where f is the frequency. The dispersion coefficient reflects the density of the passive film, which generally ranges from 0 to 1, and a large coefficient value indicates a low dispersion effect. The smaller the value of R_s is, the faster the electrochemical reaction will be. The R_ct value reflects the corrosion rate.

Table 3. Parameter values of equivalent electrical circuits.

Specimen	Rs (Ω·cm2)	Rct (Ω·cm2)	Cd (μF·cm−2)	n
untreated	555	67,227	29.47	0.79
treated	468	739,380	24.47	0.82

shows the parameters of equivalent electrical circuits in the EIS. It shows that the R_ct value of the treated surface was improved, indicating that the corrosion reaction became more difficult. For the treated sample, C_d decreased and n increased, indicating that the diffusion effect decreases and the passive film became more compact.

The corrosion resistance of the stainless steel predominantly depended on the chemical compositions and structure of the passive film, as well as the surface defects. The corrosion resistance was increased with the increase in the Cr concentration in the passivation film. The USRP reduced the grain sizes and eliminated the microscopic defects, resulting in an improved compactness and uniformity of the passive film. The corrosion resistance of the material was improved. On the other hand, residual compressive stress could be generated after the USRP due to elastic and plastic deformation during processing. The residual compressive stress could inhibit the initiation and early propagation of surface cracks, so the corrosion resistance of the metal was improved [7,16,19,28].

3.4. Fouling Performance

The fouling deposition rate was expressed by the weight gain per unit area. For the USRP-treated samples, the fouling weight per unit area was calculated indirectly through that of the untreated samples because only one surface was treated. The focus of this study was on the anti-fouling feasibility of the USRP treatment, so the calculation accuracy was acceptable. Figure 7 shows the fouling deposition weight on the untreated and treated surfaces over 30 h. The fouling on the treated surfaces was much less than that on the untreated surfaces. In the late stage, the fouling deposition rate accelerated because CaCO₃ crystals easily attached and grew on themselves.

Figure 8 presents the fouling morphologies on different surfaces after 20 min for the fouling experiments. It was found that there was less crystallization fouling that adhered to the treated surfaces, as shown in Figure 8a,c. Compared with the untreated surfaces, the surface energy was reduced due to the larger contact angle, which reduced the adhesion of fouling. There is no difference in the crystal shapes of CaCO₃ on the two surfaces, which were mainly discal and cuboidal, as shown in Figure 8b,d. In addition, it is found that fouling tended to obviously nucleate and accumulate around surface defects. This was due to the crystal surface defects resulting from surface defects, which could lead to a partial imbalance in the electrostatic interaction [21]. The negative ions and cations in some solutions were preferentially adsorbed onto the defects, resulting in increases in the concentrations of Ca⁺ and ${\mathrm{CO}}_3^{-2}$. Therefore, the partial supersaturation of the crystal species led to prior nucleation at the defects. As shown in the high-magnification images of Figure 8b,d, it was also found that CaCO₃ crystals formed in discal aragonite and cuboidal calcite. Calcite had the best thermodynamic stability and the lowest solubility among the three crystal types of CaO₃. Thermodynamically, aragonite was easily transformed into calcite. Calcite had stronger adhesion and was not easily washed away by water.

After the experiment, the samples were ultrasonically cleaned for 1 min. The fouling on the two different surfaces was removed, as shown in Figure 9. The chemical compositions of the treated sample after fouling removal was measured via energy-dispersive spectroscopy (EDS). As shown in Figure 10, the element Ca was not detected on the good surface or the defect, which proved that the fouling was cleanly removed. For the treated sample, the adhesion force may have decreased between the fouling and substrate because of the smooth surface and low surface energy. This would make it easy to remove the fouling. However, this phenomenon was not observed in this experiment, which was mainly due to the lower adhesion of fouling on the surfaces.

The above research shows that the USRP-treated surface showed better anti-fouling performance. On the one hand, the smoother surface reduced the initial deposition rate by providing fewer sites for crystal nucleation. On the other band, the larger contact angle and better corrosion resistance reduced the nucleation rate. Studies have shown that the tensile stress required to break the fouling layer away from a rough surface is several times greater than that for a smooth surface [29]. The blunter the contour peak is (as shown in Figure 1b), the more difficult the fouling is to nucleate at the peak or trough of the wave (as shown in Figure 8c). The peaks can increase the surface chemical activity, while the valleys provide a place for the sediment to avoid being scoured by fluid [21]. So, a smoother surface has a longer induction period for fouling. A layer composed of a corrosive or oxidative product, which is called a “transitional interface”, which works as a “bridge” to connect the matrix and fouling, is easily formed on metal surfaces that are easily corroded and oxidized, which increases the tendency toward deposition of calcium carbonate [14,30]. It is difficult to form a “transitional interface” on a USRP-treated surface because of the improvement in the corrosion resistance, thus increasing the difficulty of fouling deposition.

4. Conclusion

The USRP was used to modify 304SS. The physical, chemical, and fouling adhesion properties of USRP-treated surfaces were studied. A refinement layer with a thickness of about 10 μm near the surface was obtained with the USRP, which was able to improve the compactness of the passive film. The peaks and valleys of the treated surface were obviously blunter, and the surface roughness R_a was reduced from 0.19 to 0.014 μm. The water contact angle was improved from 72° to 100°, indicating the higher hydrophobicity of the metal surfaces. The USRP-treated 304SS specimen resulted in better corrosion resistance with more positive E_corr, lower I_corr, and higher R_ct values. The crystal structures on the treated and untreated surfaces were discal aragonite and cuboidal calcite. The fouling weight on the treated surface was about 0.4 times that on the untreated surface after a 30 h test. The USRP treatment was able to reduce the CaCO₃ nucleation locations and fouling deposition area by blunting the peaks and reducing the surface defect sizes. At the same time, it was difficult for fouling to be deposited and grow on the treated surface because of the better hydrophobicity and corrosion resistance.

Despite these efforts, further studies are needed, such as studies of the processing parameters. The optimal processing parameters for anti-fouling will be studied because of processing parameters in the USRP greatly affect the surface properties. In addition to CaCO₃, more chemicals in fouling also need to be considered in future studies.