Influence of surface cleaning by tools made of copper alloys on the painting performance.
Keywords Paintings, Copper tools, Surface treatment, EIS
Surface preparation is one of the most important factors in guaranteeing the expected performance of anticorrosive painting systems. It includes surface contaminant removal (oils, waxes, corrosion products, aged non-adherent coating, etc.) and creation of a suitable roughness profile to obtain a good mechanical adherence of the painting system.
Among the surface cleaning methods by mechanical action, abrasive blasting is one of the most effective--not just to achieve contaminant removal, but also to create roughness on the substrate. The high-pressure hydroblasting (> 170 MPa) is also very effective in contaminant removal, but does not improve anchorage profiles. Nevertheless, these two methods cannot always be used in maintenance painting services for several reasons. In these cases, cleaning by mechanical tools (needle gun, rotary steel brush, etc.) is widely used, although the cleaning pattern is not as good as that obtained by one of the previous methods, which reduces the painting durability.
The traditional mechanical tools are made of ferrous alloys with appropriate mechanical and corrosion properties for withstanding the service conditions. During the cleaning process, a generation of sparks is possible, especially when working with steel tools. This can be dangerous in areas containing flammable products, for example. Therefore, the use of sparkless tools is preferable in many situations, although they are less efficient. In this context, one alternative is the use of mechanical tools made of copper alloys.
When the cleaning tools are made of brass, the surface shows a yellowish aspect due to the copper alloy incrustation. Because the contact between brass and carbon steel results in a galvanic couple, one question arises: do copper alloy residues, even in small quantities, affect the painting performance? Information regarding this subject is of great importance in the anticorrosive painting field. Therefore, the aim of this paper is to evaluate the performance of a traditional painting system applied on carbon steel cleaned with mechanical tools made of copper alloy. In addition, steel plates were cleaned by abrasive blasting followed by an additional treatment using the same mechanical tools in order to compare the effect of the residues without corrosion products in the surface on the performance of a painting system.
Carbon steel AISI 1020 plates measuring 150 mm x 100 mm were exposed to an industrial atmosphere until corrosion grades varying between C and D were evidenced, according to the Swedish technical standard classification ISO 8501-1. (1) Afterwards, the plates were degreased with organic solvents and the corrosion products were removed according to the description in Table 1. The tests were carried out in triplicate, and the mean value was presented. The groups were compared using a qualitative approach to the results.
Table 1: Description of the procedure used to remove the corrosion products from the steel Group Surface preparation Characteristics A Cleaning grade St3 Brush and needles gun made of special steel B Cleaning grade St3 Brush and needles gun made of brass C Abrasive blasting near white Steel grit as abrasives metal Sa 2 1/2, ISO 8501-1 (1) (surface roughness between 40 and 85 [mu]m) D Blasting Sa 2 1/2, followed by Brush and needles gun made of cleaning with mechanical tools special steel E Blasting Sa 2 1/2, followed by Brush and needles gun made of cleaning with mechanical tools brass
Concerning the substrates of the groups D and E, the objective was to compare the effect of the cleaning tool material on painting performance. The substrates from group C were used for comparative purposes.
After the surface preparation, the steel plates were coated with a traditional painting system for ferrous substrates prepared by mechanical and manual tools:
* One coal of aluminum polyamine epoxy-mastic was applied by conventional spray gun with a minimum dry thickness of 130 [mu]m. The epoxy-mastic is a well-known type of paint used in maintenance services, especially when cleaning by blasting is not possible.
* One coat of aliphatic acrylic polyurethane finishing paint was applied by conventional spray gun with a minimum dry thickness of 70 [mu]m.
After painting, the specimen groups were labeled using the codes A, B, C, D, and E as described in Table 1. The samples for the corrosion tests received an additional coat on the edges to avoid the premature corrosion failure in these critical regions. A scratch, with approximately 7 cm and 45[degrees] in relation to the sample's shorter dimension was done on the samples submitted to natural and accelerated corrosion tests.
Elemental analysis of mechanical tools and steel plates after cleaning was made by the energy dispersion technique (EDS) with the following objectives:
(a) characterize the materials of the commercial cleaning tools (brushes, needles, and impact manual hammer) made of copper alloys;
(b) verily if the cleaning tools have left residues adhered to the steel surface.
The open circuit potential measurement of tools and steel plates after the cleaning process was made by immersion in a 3.5% NaCl solution (pH = 7) using a saturated calomel electrode (SCE) as reference. The rest time of the sample in contact with the solution was approximately 2 min.
The painting system performance applied to ferrous surfaces with different cleaning grades was analyzed by accelerated and nonaccelerated corrosion tests, and by the electrochemical impedance technique. The performance evaluation was made regarding the accelerated and nonaccelerated corrosion tests, essentially by using the criteria established in the standard ISO 4628. (2) The parameters analyzed were blistering (ISO 4628/2). corrosion (ISO 4628/3), cracking (ISO 4628/4). and peeling. In addition, the undercoating corrosion around the scratch was determined. Adherence tests were also made, referring to the initial condition and the end result of the corrosion tests. The methods used were: ASTM D 3359 [(X-cut).sup.3] and ASTM D 4541 (pull-off test). (4)
The pull-off test was carried out in a Pneumatic Adhesion Tensile Testing Instrument (PATTI) using a 1/2" pull stub.
The corrosion tests were: cyclic corrosion test (2800 h), humidity chamber (2800 h), outdoor exposure (11 months) and immersion tests in 0.6 M NaCl (2800 h) and 0.01 M NaCl (1500 h).
In the cyclic corrosion test, carried out in the laboratory, the coated samples were exposed to the following conditions:
* Seven days of exposure to ultraviolet radiation (UV-B) and humidity condensation, according to ASTM G154, (5) using the cycle of 8 h of UV-B radiation and 4 h of humidity condensation.
* Seven days of continuous exposure in the salt spray chamber, according to ASTM B 117.(6)
The exposure test in the humidity chamber was based on the standard ASTM D 2247(7) and performed with a relative humidity of 100% at 40[+ or -] 2[degrees]C.
The outdoor exposure test (field test) was carried out in the atmospheric corrosion station of CEPEL (Fundao Island, Rio de Janeiro), which has a corrosiveness category equal to C3, according to the standard ISO 9223. (8) However, to accelerate the corrosion process during the exposure, the samples were pulverized with sodium chloride solution (1% Nacl) twice a week.
The immersion tests were carried out in 0.6 M NaCl and 0.01 M NaCl. The test in the 0.6 M solution was conducted according to ASTM D 1308(9) at 40[+ or -]2[degrees]C. The immersion in the 0.01 M solution was done at room temperature. The performance of the painted samples in this test was monitored by electrochemical impedance.
The electrochemical impedance measurements were carried out using a three-electrode classic cell. The equipment used was, basically, an Omnimetra PG-19 potentiostat and a Solartron 1250 transfer function analyzer controlled by computer. The measurements were made with potentiostatic control at the open circuit potential with a perturbation of 10 mV. The frequency range measured was from 40kHz to 1.6 mHz.
The elemental analysis of mechanical tools and steel plates is presented in Table 2. As can be seen, the brush, hammer, and impact manual hammer are made of alloys containing copper and zinc. The needle is made of an iron-copper (Fe-Cu) alloy and the zinc (Zn) is absent. As was expected, the steel needle had alloy elements (chronium, molybdenum, and silicon) that are crucial--not only to improve the steel mechanical properties but also to increase the corrosion resistance.
Table 2: Results of the semiquantitive chemical analysis by energy dispersion spectroscopy (EDS) of the metallic materials and steel plates surfaces, after cleaning by mechanical tools Material Chemical elements Cleaning materials of mechanical tools Steel needle Fe (98.5%): Mo (0.8%); Cr (0.4%); Si (0.3%) Brass brush (a) Cu (66.2%); Zn (33.8%) Brass needle (a) Cu (98.3%); Fe (1.4%); Cl (0.3%) Impact manual hammer (a) Cu (63.4%); Zn (30.3%); Pb (5.5%), Ni (0.3%) Steel plates surface after cleaning by mechanical tools StA Fe (100%) StL Fe (57.8%); Cu (26.4%); Zn (15.8%) (a) After the elemental analysis, it was noticed that the alloy did not contain tin (Sn), one of the constituents of the brass
The presence of Zn and Cu on the steel surface, after cleaning with mechanical tools, is evidence of the incrustation generated by the tools, mainly for the samples from group B (treatment with mechanical tools containing brass elements). Consequently, this assures us, as expected, that this cleaning procedure causes incrustation of the mechanical tools' cleaning elements. Another aspect that confirms the analysis is the fact that the cleaning elements yellowed the surface due to the copper alloy incrustation.
Open circuit potential measurement
The open circuit potential measurements in 3.5% NaCl solution are presented in Table 3. As can be observed, all the mechanical tools, including the brush and the steel needle, have more noble electrode potential than the steel substrate (group C: -600 m[V.sub.SCE]. Thus, the incrustation of these materials on the carbon steel surface tends to form a galvanic couple; this is well characterized by the potential values obtained for the plates after the surface cleaning, mainly for the samples from groups D and E. Thus, as expected, the tools made of copper alloys tend to bring the potentials to more noble values. For the samples from groups A and B, the more noble potential values cannot be attributed exclusively to the materials' incrustation, but also to the presence of an oxide layer that remains on the surface--characteristic of the St3 cleaning grade (groups A and B).
Table 3: Results of the open circuit potential measurement Material Potential (mV)SCE Mechanical tools metallic materials Steel needle -400 Copper alloy needle -270 Rotary brass brush wire -320 Rotary steel brush wire -450 Steel plates after surface cleaning SaL -500 SaA -550 StL -415 StA -440
Cyclic corrosion test
The cyclic corrosion test
The cyclic corrosion test performed in the laboratory lasted 2800 h and the results are evaluated below:
* Concerning blistering, corrosion, cracking, and peeling, no modification was observed.
* Regarding the undercoating corrosion around the scratch, the results are presented in Fig. 1. The undercoating corrosion value is a measure of painting resistance to the progress of corrosion underneath the painting; this is an important feature to be analyzed because sometimes the coatings are damaged in service and a greater resistance to corrosion progress is desirable. Groups D and E demonstrated the greatest under coating corrosion; however, no significant difference between them was evidenced. The A, B and C groups presented similar behaviors. The difference between groups A and B was also not significant.
[FIGURE 1 OMITTED]
Humidity chamber test
The exposure test in the humidity chamber lasted 2800 h and the results are evaluated as follows:
* Concerning corrosion, cracking, and peeling, no modification was observed for any systems.
* Regarding blistering, only group C (abrasive blasting + painting system) did not exhibit the formation of blisters. The results are indicated in Table 4.
Table 4: Blistering classification for the exposure test in humidity chamber System Blisters classification SaA 2 (S2) SaL 3 (S2) StA 4 (S4)/3 (S2) StL 4 (S4)/3 (S2)
Immersion test in sodium chloride solution (3.5% NaCl)
The immersion test in sodium chloride solution (3.5% NaCl) was used to simulate exposure to Sea water. It lasted 2800 h and the results are evaluated below:
* Considering cracking and peeling, no modification was observed for any systems.
* Regarding corrosion and blistering, the C, D, and E groups did not exhibit any evidence of them. However, the A and B groups presented modification concerning corrosion and blistering. The results are presented in Table 5.
Table 5: Blistering and corrosion classification for the immersion test in sodium chloride solution System Blistering Corrosion StA 2 (S2) Ri 2/Ri 3 StL 4 (S2)/3 (S2) Ri 1
Outdoor exposure test
The outdoor exposure test (field test) was carried out in the atmospheric corrosion station of CEPEL (Fundao Island, Rio de Janeiro), which has a corrosiveness category equal to C3, according to the standard ISO 9223, which is a sea-side situation. It lasted 11 months and the results are presented below:
* None of the samples presented blistering, cracking, or peeling.
* Considering corrosion, all systems except for group A demonstrated corrosion rust spots on the coatings surface, which were classified as Ri 2 (0.5%).
* Concerning undercoating corrosion around the scratch, the result are shown in Figs.2 and 3. Similar to the cyclic corrosion test, groups D and E presented greater undercoating corrosion; however, group E demonstrated better performance than group D. Groups A and B exhibited similar behavior and group c presented the lowest undercoating corrosion.
As previously described, adherence tests were carried out before and after the corrosion tests. Concerning the results obtained by the X-cut method (ASTM D 3359), all systems presented the 5A classification before and after the corrosion tests. The pull-off test (ASTM D 4541) results are shown in Table 6. In this case, the type of adherence failure was analyzed in addition to the rupture force.
Table 6: Adherence tests results by the pull-off test, referred to initial and after the corrosion tests conditions Initial After cyclic After condition test (UVB- immersion salt spray) in NaCl 3.5% solution Group MPa Type of MPa Type of MPa Type of failure failure failure C 6.0 B 5.6 C 9.4 C D 4.0 B 2.5 75%C 3.1 C 25% Y/Z E 6.0 B 2.9 75% Y/Z 3.6 C 25% C A 4.5 60%C/Y 2.9 60% Y/Z 6.5 C 40% C B 5.0 B 4.8 75% A/B 2.9 A/B 25% C After exposure After feild in humidity test (11 months) chamber (100%) Group MPa Type of failure MPa Type of failure C 10.0 C 4.3 Y/Z D 2.2 50% A/B 3.4 Y/Z 50% C E 4.0 C 2.4 Y/Z A 6.4 90% C 4.5 Y/Z 10% A/B B 3.5 A/B 4.4 Y/Z
The pull-off test consists of fixing a pull-stub on the coating by means of an appropriate adhesive, and screwing the stub on a piston to pull it apart from the coating. The piston is attached to a PATTI, Which applies the pressure necessary to completely debond the stub. The derive measures this pressure and the failure location is classified. The substrate specimen is designated as A, upon which successive coating layers designated as B and C have been applied. The adhesive is designated as Y and the stub as Z, as mentioned in ASTM D 4541. In Table 6. the cohesive failures are designated by the layers within which they occur as A, B or C. and the percent of each. The adhesive failures are designated by the interfaces at which they occur as A/B, B/C, C/Y, or Y/Z, and the percent of each.
As can be noticed, the results obtained by the X-cut method were excellent, not only before but also after the corrosion tests; thus, the different mechanical tools materials did not show any influence on the performance of the tested systems. Considering the lest carried out by the pull-off test, as Table 6 shows, a slight superiority of group A to B can be observed in the immersion tests in 3.5% NaCl solution and in the humidity chamber, in which the nature of the failure was A/B type (substrate/coating).
For the samples with St3 cleaning grade, the remaining corrosion layer can be structurally heterogeneous; therefore, the adherence results must be carefully analyzed to avoid incorrect conclusions. This means that the results can be related to another factor instead of the metallic materials incrustation (steel and copper alloys). This can be evidenced by the adherence of groups D and E. in which the plates were free from corrosion products. From a qualitative point of view. no significant differences were observed in most of the tests. After the humidity chamber test, group E presented a better result than group D.
Electrochemical impedance spectroscopy
The resistances of the painting systems in the function of immersion time obtained by electrochemical impedance measurements are presented in Fig. 4. The impedance diagrams were characterized by only one loop. The resistance values were obtained by extrapolating the loop to the real axis. The systems with St3 cleaning grade (groups A and B) presented lower resistances than those in which the surface preparation was made by blasting (groups C, D, and E). No significant difference relative to the mechanical tools materials (steel and copper alloys) in this parameter was observed.
By the energy dispersion analysis, it can be observed that the cleaning procedure with mechanical tools caused the metallic materials incrustation (steel and copper alloys) on the samples surfaces. It is worth emphasizing that this is expected in the cleaning process by mechanical action. Even for cleaning by blasting, the abrasive material incrustation occurs, as mentioned in a previous work. (10) Due to the more noble potential of the tools materials compared to the substrate (steel), the incrustation causes the formation of a galvanic couple. as can be detected by the results presented in Table 3. A greater potential difference, as expected, was observed for the samples cleaned with copper alloys tools. Therefore, the initial existing doubts, related to the influence of the superficial incrustation on the paintings' performance, were really conceivable--thus, it is worthwhile to study this influence.
For the cyclic corrosion test, no significant influence of the copper alloy incrustation on the paintings' performance was detected--not only in groups D and E, but also in groups A and B. For the areas away from the scratch, all the systems presented a similar performance and were free from failures such as corrosion. blistering, peeling, and cracking. Around the scratch, the undercoating corrosion was greater for groups D and E. However, the difference was very small and insufficient to make conclusions regarding the tool materials" influence. The greater undercoating corrosion observed for groups D and E, compared to the others (mainly to group C), can be attributed to the use of mechanical tools after blasting. Using scanning electronic microscopy, it was observed that this procedure substantially reduced the roughness of the surface. This factor certainly contributed to the greater undercoating corrosion around the scratch. Comparing groups A and B. the performances around the scratch area were very similar and the observed difference is not enough to characterize the influence of the mechanical tools' material on the paintings performance.
For the exposure test in the humidity chamber, again, no influence of the tools' materials on the paintings performance was observed. Group C did not present any type of failure; the others (groups A, B, D, and E) presented blisters in the coating. Considering the adherence tests using the pull-off test, they did not demonstrate a clear influence of the cleaning tools materials on the paintings performance.
Concerning the immersion test in sodium chloride solution, a poor performance was observed for groups A and B. since both of them presented blisters and corrosion. These results were expected since it is known that the paintings performance depends primarily on the surface cleaning grade. However, the obtained results did not evidence any influence of the cleaning tools' materials on the paintings performance; in brief, groups A, B, D, and E presented very similar behaviors. Unlike the previous tests, the adherence results did not evidence a significant influence of the cleaning tools materials on the paintings performance.
For the outdoor exposure lest, with sodium chloride pulverization twice a week, no deleterious effect of the copper alloy tools on the paintings performance was observed. In spite of the same value of undercoating corrosion, group A presented a poorer performance than group B. After 11 months of exposure, the first one had already presented corrosion spots on the steel surface, as can be seen in Fig. 3.
Around the scratch area, no deleterious effect of the brass incrustation on the paintings performance was observed. As can be noticed in Fig. 2, groups A and B presented the same behavior. Among groups D and E, the first one presented an even greater undercoating corrosion. As previously described, the greater undercoating corrosion of groups D and E can be related to the reduction of the surface roughness, due to the use of mechanical tools after the abrasive blasting. Regarding the undercoating corrosion in group D's system, compared to group E, it is possible that the use of steel tools caused a greater reduction of roughness due to the greater hardness of the metallic cleaning materials. The steel needles have 48 Rockwell C of hardness and the copper alloy tools have 27 Rockwell C.
[FIGURE 2 OMITTED]
[FIGURE 3 OMITTED]
The adherence results, by both methods, were very satisfactory. They showed that there is no influence of the mechanical tools' materials on the paintings performance.
The electrochemical impedance spectroscopy results presented in Fig. 4 exclusively show the effect of the surface cleaning grade on the paintings performance. As expected, groups A and B (St3 cleaning grade) demonstrated lower resistances values; thus, a poorer performance than the others (groups C, D, and E). In the previously discussed traditional corrosion tests analysis, no influence of the mechanical tools materials on the painting performance was observed.
[FIGURE 4 OMITTED]
Therefore, despite the existence of a galvanic couple, the incrustation of copper alloy to the substrate did not impair the paintings performance (groups B and E) in comparison with the systems with surface preparation by steel tools (groups A and D). Such behavior can be attributed to the painting system that, with a high resistance, avoids the formation of galvanic couples, and to the fact that the anodic area is greater than the cathodic area.
It is important to highlight that the mechanical tools with cleaning materials made of copper alloys (e.g., brass)--despite the fact that they do not generate sparks during the surface preparation process--present a much lower efficiency, in terms of productivity, when compared to those with cleaning elements made of special steels. The materials' hardness is, certainly, one of the factors that directly influence the cleaning rate, because the mechanical tools made of copper alloys have lower hardness.
* The cleaning of the carbon steel surfaces by mechanical tools made of copper alloys does not impair the painting system performance, in comparison with cleaning by mechanical tools made of special steels.
* As expected, blasting (Sa) was the surface treatment that provided the best performance of the studied painting systems, and the systems with St3 cleaning grade (groups A and B) provided the worst performance.
(1.) ISO 8501, Preparation of Steel Surfaces Before Application of Paints. ISO, Geneve (1988)
(2.) ISO 4628--Parts 1-5, Paint and Varnishes--Evaluation of Degradation of Paint Coatings--Designation of Intensity, Quantity and Size of Common Types of Defect. ISO, Geneve (1984)
(3.) ASTM D 3359, Standard Test Methods for Measuring Adhesion by Tape Test. ASTM (2002)
(4.) ASTM D 4541, Standard Test Method for Pull-Off Strength of Coatings Using Portable Adhesion Testers. ASTM (2002)
(5.) ASTM G 154, Operating Fluorescent Light Apparatus for UV Exposure of Nonmetallic Materials. ASTM (2001)
(6.) ASTM B 117, Standard Practice for Operating Salt Spray (Fog) Apparatus. ASTM (2007)
(7.) ASTM D 2247, Standard Practice for Testing Water Resistance of Coatings in 100% Relative Humidity. ASTM (2002)
(8.) Morcillo. M, et al., Corrosao e Proleccao de Metais nas Atmosferas da Iberoamerica, Parte I. CYTED. Madrid (1988)
(9.) ASTM D 1308, Effect of Household Chemicals on Clear and Pigmented Organic Finishes. ASTM (1987)
(10.) Fragata, FL, Meduna, JR. "Estudo de Novo Tipo de Abrasivo a Base de Oxido de Alumfnio ([alpha]-[Al.sub.2][O.sub.3])." 17[degrees] Congresso Brasileiro de Corrosao, ABRACO, Vol. II, Rio de Janeiro. Outubro 1993
F. L. Fragata. C. C. Amorim, A. P. Ordine
CEPEL, Eletrobras Research Center, P.O. Box 68007, CEP 21944-970 Rio dc Janeiro, Brazil
M. C. Marroig, R. O. Mota, I. C. P. Margarit-Mattos
LNDC-Laboratory of Non-Destructive Tests and Corrosion "Prof. Manocl de Castro", EE/PEMM/COPPE, Federal University of Rio de Janeiro, P.O. Box 68505, CEP 21945-970 Rio dc Janeiro, RJ, Brazil e-mail: firstname.lastname@example.org
I. C. P. Margarit-Mattos
Department of Inorganic Processes, School of Chemistry, Federal University of Rio de Janeiro, Rio dc Janeiro, Brazil
|Printer friendly Cite/link Email Feedback|
|Author:||Fragata, F.L.; Amorim, C.C.; Ordine, A.P.; Marroig, M.C.; Mota, R.O.; Margarit-Mattos, I.C.P.|
|Date:||Jun 1, 2009|
|Previous Article:||Application of polyaniline/nylon composites coating for corrosion protection of steel.|
|Next Article:||Effect of heat treatment at 1300[degrees]C on W coating prepared by double-glow plasma on carbon/carbon composite.|