Test Methods for the Evaluation of Protective Organic Coatings

Please download to get full document.

View again

of 10
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Information Report
Category:

Documents

Published:

Views: 2 | Pages: 10

Extension: PDF | Download: 0

Share
Related documents
Description
articulo tecnico
Transcript
  D ~ronm n ~. d Elect mlcal st eth r the E ion f Protective Org ni o~tings 8,S. Skerry, A. Alavi, and I. Lindgren The Sherwin-Williams Company' Three accelerated test chamber methods and two electrochemical test methods have been used to evaluate anticorrosion properties of organic coat-ings. Standard salt spray (ASTM B 117) procedures were modified to include the effects of wet/dry cy-cling, various electrolyte specie NH4 +, Na +, S04 2 - and CI-), and UV radiation. Different modes of deterioration were observed depending on the test conditions used. Degradation and corrosion were studied by scanning electron microscopy (SEM)/ x-ray analysis (EDX) methods. Brief comparisons are made with data from exterior exposures at marine and industrial test sites. In separate experiments, electrochemical data were obtained for protective coatings under fully immersed conditions (0.6 M NaCI) by using ac impedance and electrochemical noise monitoring techniques. Impedance data are open to mechanistic interpretation, whereas the volt-age and current noise signatures give clear indica- tionsas to the state of the coating. iNTRODUCTION Organic coatings are commonly used to protect metallic engineering structures from degradation caused by natural or artificial environments. A recent estimate suggests that in 1986, the total value of shipments of paint and allied products in the U. S. amounted to some 10 billion. 1 It has been suggested that half of the value assessed in this way is for coatings used for corrosion protection. 2 Despite the large market for organic coatings, there are no laboratory test methods available which can be used Presented at the 65th Annual Meetlng of the Federation of Societies for Coatings Technology, n Dallas, TX, on Uctober 5 1987. 10909 S. Cottage Grcne A\e., Chicago, [L60628. ournalof unambiguously to predict the long-term protective capabilities of a new corrosion protective coating. Particularly for coatings intended for atmospheric service, heavy reliance is often placed on essentially subjective results obtained from routine test methods, such as continuous salt spray. The most widely used test in the U.S. for this purpose is the procedure described by ASTM Method B 117-85. 3 The justification for the use of such a method is that if a coating system resists the aggressive conditions of the salt spray, then it will also resist subsequent service environments. However, the non-justified assumption being made is that the mechanisms of corrosion and degradation pertaining to the service environments will be the same as those in the specific conditions of the continuous salt spray test. In fact, it has been recognized for many years that when ranking the performance levels of organic coating systems, there is little if any correlation between results from standard salt spray tests and practical experience. 4-6 Consequently, the need to develop more useful accelerated laboratory test methods for the evaluation of protective organic coatings is an important current issue for the coatings industry. Describing meaningful accelerated testing procedures for coatings intended for atmospheric service is not a trivial task considering the complexities of the natural environment. Factors which affect corrosion and degradation of coated metallic substrates will certainly include the relative availability of oxygen, moisture, sunlight energy, and ambient temperatures, as well as the chemi cal content of the atmosphere (e.g., 03, S02, NO x) and also of alighting rainwater (e.g., H+, Na+, Ca 2 f , NH4 + , Cl- , SO/ -, N0 2 . Technol rint  B.S. SKERRY, A. ALAVI, and K.1. LINDGREN 7 llb~<i:l 1 COSJ~Dl1g\li Slff l~l/lm\li Studierdlby l@l:»tH©I~orlf Accelerillted 'fest Methods Alkyd (active pigments: zinc chromate zinc oxide) iEpoxy-polyamide (active pigment: zinc phosphate) AplProx. Totall ~y film Thickness TOpcoSlt SubstrSlle (fLm) _ Alkyd Mild steel 70 (SAE 1010; 0.5 1.0 J..lm pro-file, washed by d.i. water MEK) Epoxy-polyamide Blast steel 135 (cold rolled; 37-50 f1ITI profile, washed by d.i. wateriMEK) Although it is not the purpose of the present work to mode the complexities of the natural environment, three clear factors of importance in corrosion/degradation studies which are not modelled by the standard (ASTM B 117-85) salt spray test have been studied. They are: wet/ dry cycling, a suitable choice of electrolyte, and the effects of incident UV radiation. Furthermore, since metallic corrosion is an electrochemical phenomenon, it is believed that electrochemical test methods will provide a source of quantitative performance data for coatings evaluation work. For example, ac impedance methods have been investigated extensively for this purpose and considerable progress has been reported. 7- 10 For this reason, ac impedance has been used in the present work to obtain information regarding the performance capabilities of anticorrosion primers. A relatively new approach utilizing electrochemical noise monitoring has also been studied. The methods and the rationale for the electrochemical noise work have been described in more detail elsewhere. II For completeness, brief details of these techniques are also given in this paper. iEl'ii'0'~r.tl)'lIm@Uit21~ T®§t Fll'oc@dllJ1utes LABORATORY ACCELERATED TEST METHODS: Three laboratory accelerated test methods have been studied by exposing coated mild steel (6 x 3 ) test panels to the environments described in this section. Test panels were prepared under standardized conditions using numerous coating systems (primers with and without topcoats) on a variety of substrates. In this paper, results are given for the coating/substrate combinations described in Table 1 Scanning electron microscopy (SEM) combined with energy dispersive x-ray analysis (EDX) methods were used to examine corrosion and degradation processes on exposed test samples. These studies were conducted using a Cambridge Scientific Instruments Stereoscan 250 fitted with a Tracor Northern TN 2000 x-ray analysis spectrometer [142 eV full width, half maximum peak resolution based on the Fe KOL) peak.] The three accelerated test methods studied are described in the following sections. For each test condition, the total exposure time used was a maximum of 2000 hr. Standard Salt Spray Test ASTM B 117-SJl)-In this procedure, coated panels were exposed inside a standard test chamber, supported at a 15-30° angle from the vertical, and were sprayed indirectly with an atomized 5 wt% salt spray based solely on NaC . The test was operated at 35 ± I.5 0 C with a steady state relative humidity of approximately 95-98% Wet Dry Cycle Corrosion Test Variations on the standard salt spray test method have been in existence for many years. Some of these are described in ASTM G 85-85. 12 Elsewhere in the literature, Harrison, Tickle, and Barraclough 4 13 concluded that salt spray tests based on continuous NaCl spray alone are particularly unreliable for accelerating atmospheric corrosion in industrial atmospheres. These workers suggested that the observed unreliability was because. of the absence of ammonium and sulphate species in the test method and also because the effects of wet/dry cycling which occur in nature are not reproduced in continuous salt spray conditions. They clearly demonstrated the presence of both sulphate and ammonium species both in the atmosphere as well as on naturally weathered steel. Accordingly, Harrison 14 used a salt spray solution comprising 3.25 wt% (NH4 hS04 with 0.25 wt% NaCI, and observed improved correlations with 14-year exterior exposures in industrial environments. These ideas were further developed by Timmins 15 who recommended using wet/dry cycling 3 hr salt spray alternating with 1 hr drying using ambient air) and also the use of a diluted version of Harrison's electrolyte, comprising 0.40 wt% (NH4 hS04 with 0.05 wt% NaCl. The wet/dry cycle test used in the present work is similar to that described by Timmins 15 The method used was a wet/dry cycle conducted in a test chamber approximately 0.32 m 3 in volume. Timmins' electrolyte was used, having an approximate pH of 5.2. Salt mist was produced by a venturi nozzle atomizer at a rate of 600 mL hr I. All coated panels received a similar quantity of electrolyte when located around the periphery of the chamber. During the wet cycle, the chamber was allowed to cool under ambient conditions. In this work, dry and wet cycles were programmed for 1 hr each in order to accelerate further the effects of wetting and drying compared with Timmins' method. During the dry cycle, the test chamber was resistively heated to 35 ± 2°C. Thus, the coated panel surfaces appeared visibly dry at the end of each drying cycle. The reiative humidity inside the test chamber was reduced to approximately 45% at the end of each dry cycle as measured by a wet/dry bulb hygrometer. A further point to note is that with the electrolyte concentrations used in this test (substantially lower than in conventional salt spray) the corrosion inhibitive pigments may function more as they would in a natural environment; an issue already raised in the literature by Funke. 16  WetlDrylUV Condensation Test UV degradation and condensation factors were incorporated into the wet/dry cycle corrosion test method prcviously described by making use of a standard UV -condensation test apparatus operating to ASTM G 53-84 specifications. 17 In this work, the conditions used were: 4 hr radiation (UV -B 313 bulbs) at 60 D C followed by 4 hr condensation deionized water) at SODe. This test procedure was intended to simu late deterioration caused by water as dew and also the concomitant effects of natural UV radiation. In this third test method, test panels were exposed to the wet/dry corrosion test method for 200 hr, then cycled through thc UV -condensation test chamber for 200 hr on a rotating basis until a total of 2000 hr had elapsed. OUTDOOR EXPOSURE TESTS: In the present work, some initial comparisons have been made with results obtained from outdoor exposures. Two experiments were conducted in order to study the morphology of corrosion products produced in natural environments, and also the chemical composition of corrosion products formed in scribe lines of painted panels exposed at industrial and marine test sites. In the first of these experiments, uncoated mild steel panels (SAE 1010:C 0.08-0.13%, Mn 0.3-0.6%, P max) 0.04%, 0.05 having a surface finish roughness of approximately 0.5-1.0 fJ. m were exposed for four weeks to an industrial atmosphere (Chicago, . Corrosion produc[ morphology was studied by SEM. The results obtained were then compared with the corrosion products formed on similar mild steel panels after 48 hr exposure to: a) salt spray and b) wet/dry cycle corrosion test conditions. In the second experiment, the corrosion products from the scribe lines of two coated panels after exterior exposures for six months in natural environments were studied by EDX. Corrosion products were analyzed from the coated panels described in Table 2. Two electrochemical test methods have been used in the present work. are, firstly, ac impedance, and secondly, an analysis of the voltage and current noise transients which occur under intact paints when under freely corroding test conditions. The experimental test procedures used have been described in detail elsewhere. 11 For both techniques, the test electrodes used were coat ed mild steel (SAE 10 10) test panels of surface area either 24 cm 2 or 40 cm 2. Plexiglass cells were attached to these panels which were filled with 0.6 M NaCI as the electrolyte. Ambient aeration was used during testing. AC IMPEDANCE: All ac impedance mcasurements were conducted using a Solartron 1250 frequency response analyzer FRA) operated under microcomputer control HP 85). The FRA was connected to the electrochemical cell via a Thompson 251 potentiostat. For studies of relatively ow impedance systems, a standard potentiostatic three electrode configuration was used with the coated working) electrode held at the measured free corrosion potential. For relatively high impedance systems, a two ENVIRONMENTAL AND ELECTROCHEMICAL TEST METHODS Table C©lSl~@d Pai1els iElqoosed Outdoors 10r ubsequen~ ~ [:rifbl@ line Cmmsion Product AI1Sllysis Approl1. Total Dry Film ·rl1iclmess El1posure rimer Topcoat Substrate (f.lm) Environment Latex None Mild steel 100 Marine active SAE 1010; Florida) pigment: 0.5-1.0 fLm barium profile, metaborate) washed by d.i. water/MEK) Epoxy-Poly-Mild steel 85 Industrial polyamide urethane SAE 1010; Chicago) active 0.5-1.0 fLITI pigment: profile, zinc phosphate) washed by d.i. water/MEK) electrode configuration was used with the FRA operating in its amplitude compression mode. Applied signals were in the range 20-50 mV over a nominal frequency range of 10 kHz-10 mHz. Representative complex plane data are given for the primer coatings noted in Table 3 ELECTROCHEMICAL NOISE ANALYSIS: For the electrochemical noise monitoring experiments, electrochemical cells previously described were prepared and assembled in a three electrode arrangement wherein two of the electrodes were coated steel substrates prepared so as to be nominally identical. These two coupled electrodes were connected electrolytically by an agar salt bridge and allowed to corrode freely during a period of approximately 2000 hr. The third electrode in the assembly, in each case, was an Ag AgCl reference. Periodically during testing, a low noise zero-resistance ammeter ZRA) was connected between each coupled electrode pair. Two sensitive digital voltmeters (DVM's) measured, simultaneously, each cell current flow via the ZRA) and each pair electrode potential. The experiment was conducted under microcomputer control HP 85) using eight separated channels multiplexed with a commercially available IIEEE programmable switch. Data were collected as time records of coupling currents and potentials for each pair of coated electrodes. By applying an Ohm's Law analogy as discussed previous ly 11 derived values of an approximate' 'polarization re sistance parameter Rp) for each coated electrode pair were obtained from the simple ratio of the standard deviation of the potential noise signal to the standard devi- T<OIble 3-Primer Coatings Studhed blf Ac Impec smce Approx. Dry Film Primer Vinyl chloride barrier type) Alkyd active pigment: zinc chromate) Substrate Mild steel SAE 1010; 0.5-\.0 fLm surface profile, washed by d.i. water/MEK) Mild steel SAE 1010; 0.5-1.0 fLm sufacc profile, washed by d.i. water/MEK) Thickness (f.l m) 60 30  B.S. SKERRY, A. ALAVI, and K.I. LINDGREN u@iJi@ 4---CI J< lItilfl ll' Stll llii:li~ill illy lEi~~~f@([;U1®mic;aJ~ Nense Anailf$is AppmN.DI'jf film Thickl1l llss JLm) oatil1@ Polyurethane (barrier type) Epoxy-polyamide (active pigment: strontium chromate) Alkyd (barrier type) Alkyd (active pigment: zinc chromate) Mild steel (SAE 1010; 0.5-1.0 fLm surface profile, washed by d.i. water/MEK) Mild steel (SAE 1010; 0.5-1.0 fLm surface profile, washed by d.i. water/MEK) Mild steel (SAE 1010; 0.5-1.0 fLm surface profile, washed by d.i. water/MEK) Mild steel (SAE 1010; 0.5-1.0 fLm surface profile, washed by d.i. water/MEK) 30 40 40 30 ation of the current noise signal, i.e., Rp aV/ai. It should be noted that this derived resistance parameter encompasses effects due to the coating as well as to charge transfer and diffusion processes. For coated electrodes, these effects cannot be separated ouC Representative data are presented here for four coating systems as described in Table 4 Eri'~~ i'oll1lm®nta~ u@st LABORATORY ACCELERATED TEST METHODS: Visual Observations Figure 1 illustrates the deterioration observed on the alkyd coating system after 2000 hr testing in the three accelerated test environments. It is dearly apparent that the modes of degradation and types of failure thus produced depend fundamentally on the test conditions to which the coatings have been subjected. in this example, the alkyd primer/topcoat sYStem exhibited severe degradation as a result of exposure to the standard salt spray environment after only 1000 hr testing as shown in Figure . However, the degradation observed is quite unlike that observed in practice for alkyd systems. The same paint exposed to the wet/dry cycle corrosion test exhibited relatively little degradation even after 2000 hr testing, as can be seen in Figure l b). In contrast, Figure I (c) illustrates how the additional presence of UV -condensation factors in the wet/dry cycle corrosion test fundamentally altered the nature of the corrosion/degradation processes which can occur. The results produced by this combination of test factors appear, qualitatively at least, to be more closely representative of the corrosion and degradation observed for such alkyd paints in natural atmospheric service environments. A second example, shown as Figure 2, is for an epoxypolyamide primer/topcoat system applied over blast steel. This coating shows analogous trends. Neither the salt spray test result [Figure 2(a)] nor the wet/dry cycle corrosion test [Figure 2(b)] give particularly realistic deterioration effects. Again, in contrast, the added presence of the UV -condensation factors combined with the wet/dry cycle corrosion test conditions gave more realistic deterioration for this coating system, with some rusting in the scribe lines as well as some rust staining adjacent to the scribes being visible in Figure 2( c). Also apparent was the observation that some loss of gloss had occurred after 2000 hr exposure to the wet/dry/UV -condensation test cycle combination. SEMIEDX Analysis of Coated Panels After Exposure to Accelerated Test Environments In order to elucidate the effects of the additional UV -condensation factors on the overall degradation of the epoxy-polyamide paint system, SEM/EDX methods were employed. These studies . revealed that the surface morphology of the coating was profoundly affected by the nature of the test atmosphere. Figure 3(a) shows the surface of the unexposed control sample. Relatively smooth polymeric binder can be seen in this case with the added presence of some inorganic pigments underlying the surface. The effect of the wet dry cycle corrosion atmosphere was to roughen the surface slightly, and to make more obvious the presence of the pigment particles as seen in Figure 3(b). EDX analyses confirmed that the pigment particles observed were a combination of Ti0 2 and extender material. By comparison, the additional presence of the UVcondensation factors in the wet/dry cycle corrosion test method produced very significant changes in the nature of the painted surface. Figure 3(c) illustrates that, in these circumstances, much of the surface layer of the binder lFiguw® i-Visual degradation of alkyd coating system on ~®el after eNposure to laboratory accelerated es~ conditions: (a) 1000 hr salt spray (t\STM B 117-35); (b) 2000 hr wet/dry cycle [(NH4)2S0. MaCI e eciroiyte]; and (c) :W )O 1 If wetidry/UV-com: ensaticm cycle
Recommended
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks