Particle Size Influences on the Coating Microstructure through Green

Jun 12, 2017 - Chen-Yu Liu†, Dah-Shyang Tsai† , Jian-Mao Wang†, James T. J. Tsai‡, and Chen-Chia Chou§ .... Kinsey, Slusarski, Sosa, and Weih...
0 downloads 0 Views 3MB Size
Research Article www.acsami.org

Particle Size Influences on the Coating Microstructure through Green Chromia Inclusion in Plasma Electrolytic Oxidation Chen-Yu Liu,† Dah-Shyang Tsai,*,† Jian-Mao Wang,† James T. J. Tsai,‡ and Chen-Chia Chou§ †

Department of Chemical Engineering, National Taiwan University of Science and Technology, 43, Keelung Road, Section 4, Taipei 10607 Taiwan ‡ Department of Materials Engineering, McGill University, 3610 University Street, Montreal, Quebec H3A 0C5 Canada § Department of Mechanical Engineering, National Taiwan University of Science and Technology, 43, Keelung Road, Section 4, Taipei 10607 Taiwan ABSTRACT: In an effort to color the aluminum alloy surface green via plasma electrolytic oxidation (PEO), two alkaline solutions have been employed with particulate inclusions and sodium aluminate. Electrolyte I comprises a self-made chromia pigment with a mean particle size 69 nm, whereas electrolyte II contains a commercially available pigment, GN-M, with a larger particle size 351 nm. Both pigments are oxygen deficient Cr2O3‑δ of corundum-type structure before coating, the oxidative environment of PEO converts them into stoichiometric Cr2O3. In electrolyte I and II, the oxides of chromium and aluminum deposit simultaneously under analogous PEO conditions, yet resulting in very different microstructures. The GN-M inclusion of large size amasses on top of the coating, while the self-made inclusion goes deep, and closely associates with alumina and pores. The oxide coating, grown in electrolyte II, consists of a top Cr2O3-rich layer and a dense alumina layer underneath, delineated by the boundary marked with microdischarge burns. On the other hand, the self-made particulate inclusion appears to bring the electric microdischarges inside the coating and create inner pores and damages. The structure difference, caused by the difference in microdischarge locations, is attributed to shifting of the Cr2O3−Al2O3 interface where p-type and n-type semiconductors meet. KEYWORDS: chromia inclusion, aluminum alloy, plasma electrolytic oxidation, oxide semiconductor

1. INTRODUCTION

discharges to establish the desired bonds between inclusions and coating matrix. One popular inclusion is the zirconia powder, which has been reported to enhance corrosion, hardness, and thermal shock resistances.11−18 On protecting the magnesium alloys, incorporation of ZrO2 particles deserves special attention since Mg dissolution may help stabilize the zirconia phase and toughen the coating.19−21 Another additive is the alumina powder of alpha phase, which possesses many favorable properties compared with other transition phases. Although PEO may grow alpha alumina without resorting to particulate inclusion, the alpha phase generally emerges later in coating after the gamma phase has established considerable thickness.22,23 Incorporation of the alpha Al2O3 powder is considered to assist in corrosion resistance, wear resistance, and hardness of the workpiece.24−29 Inclusion of silicon carbide particles has also been attempted. The results seem to be disappointing, because the nonoxide is partially converted into silicates in plasma.30−32

In the literature on plasma electrolytic oxidation (PEO), coloring the valve metal surface has been performed through adding either soluble metal salts1−4 or suspended pigments5 in the electrolytic solution. We prefer pigment to metal salt since the pigment is more manageable in industrial practice. Green coloration with chromium is a good example. Coloration with the chromium salt involves the waste disposal problem, since water-soluble chromium poses a health hazard to the public. On the other hand, the pigment of Cr(III) oxide is chemically inert and environmentally friendly. Thus, Cr(III) oxide is widely used as the green colorant in the paints industry, showing excellent fading resistance when exposed to sunshine and rain.6 During PEO, pigment inclusion performs in the same manner as the in situ uptake of functional particles.7 Intended particles are suspended in electrolyte, driven by the electric field, adsorbed, and taken in by the growing layer. Therefore, particle-inclusion PEO is also known as electrophoretic enhanced micro arc oxidation.8−10 Incorporation is considered critical among these steps, yet less understood, since incorporation involves complicated physical or chemical events. These physical and chemical events are energized by micro© 2017 American Chemical Society

Received: March 3, 2017 Accepted: June 12, 2017 Published: June 12, 2017 21864

DOI: 10.1021/acsami.7b03113 ACS Appl. Mater. Interfaces 2017, 9, 21864−21871

Research Article

ACS Applied Materials & Interfaces

was set 400 V in positive and 100 V in negative polarization. The frequency was 500 Hz, defined as 1/(T+on + T+off + T−on + T−off), in which T+on and T−on were the duration periods of positive and negative pulses, and T+off and T−off were the resting periods between the positive and negative pulses. The duty ratio was 70%, defined as T+on/(T+on + T+off + T−on + T−off). Growth time of the barrier layer was set to 5 min; that of green coating was 40 min. In growing the green layer, the electric current of bipolar pulsed waveform was controlled, instead of voltage. The settings were referred as the controlled current mode. The PEO operation was denoted by its current settings. For example, 2.0A(+)/2.0A(−) indicated the positive and the negative currents were set to 2.0 A, which corresponded to current density 92.2 mA cm−2. Growth of the green layer and the resulting microstructure were the main issue of this study. The influences of duty ratio were varied between 10 and 40% with a fixed frequency 500 Hz. Subsequently, the influences of positive current were investigated too. Size distribution of the dispersed pigment was measured using dynamic light scattering (DLS) technique. A 0.01 mL sample of electrolyte I or II was drawn with a micropipette, diluted into a 5 mL colloid, and placed in a DLS cell holder for measurement (Zetasizer, Nano ZS, Malvern). Crystalline phase analysis was performed with either grazing incidence angle X-ray diffraction (D8 Discover, Bruker) for surface sensitive characterization, or conventional θ/2θ diffraction (D2 Phase, Bruker) for the PEO coatings and the powder samples. The X-ray source was a CuKα radiation (0.154060 nm) and nickel filter. Morphology was examined using a scanning electron microscope (SEM, JSM-6390LV JEOL), equipped with an energy dispersive spectrometer (EDS) for elemental analysis. Adhesion qualities of several chosen coatings were measured according to the 3 M peeling test with a pressure sensitive tape 600 (3 M Brand, CID AA-113, Type I class B). The tape was 18 mm in width. We cut a strip ∼45 mm in length, which was firmly pressed on the green surface and pulled up rapidly. There was a small amount of pigment remained on the tape surface and a picture was taken for comparison.

Even though microdischarges in PEO may reach several thousand degrees Kelvin, these high-energy breakdown events are also brief in duration and restricted in size.33 It is of interest to know whether the foreign particles melt or react with the coating matrix. A study on SiO2 particle inclusion indicates the particle size plays a critical role. When the particles are nanometer in size, they are reactively incorporated into the coating. When the particles are micrometer in size, they are inert with the rest of coating.34 Particles of high refractoriness, such as ZrO2 and α-Al2O3, may not melt in plasma, yet their product compounds denote reactive inclusion. For example, the presence of MgAl2O4 spinel, along with its increasing content with coating time, indicates that α-Al2O3 particles have been taken into the coating on AM60B alloy reactively.35 Detection of a new Mg2Zr5O12 phase assures the ZrO2 particle in electrolyte has bonded with the Mg-containing species in deposition.18 Reactive uptakes of low and high melting-point particles have been compared and suggested as a measuring stick for the plasma temperature.36 In this work, we grow the PEO green coatings on 6061 aluminum alloy in the aqueous solutions of sodium aluminate and chromia pigment under a controlled current mode. The electrolytic solutions are prepared with the same solid loading of two pigments, which differ in particle size, ∼70 nm and ∼350 nm; respectively. The cell voltage tends to run away prematurely in both electrolytic solutions during operation, implying the inclusion brings intriguing changes in the electrical system. These changes are manifested in the resultant coating microstructures which are contrasted and analyzed.

2. EXPERIMENTAL SECTION Two green pigments, a self-made powder and a commercially available powder (GN-M, Bayer Chemicals), were utilized in preparing the colloidal electrolytic solution. A concentrated solution of pigment was made in the following procedure. A 300 mL colloid was balled milled, containing 20.0 g green pigment and 1.6 g dispersant (Darvan C−N, Vanderbilt Minerals). Ball milling was carried out with 5 mm zirconia balls in a plastic jar for 2 days. The colloid was then mixed with a sufficient amount of water that contained 8.0 g NaAlO2 and 0.75 g caustic soda, such that the final volume was 2 dm3. We labeled the electrolyte with self-made pigment as electrolyte I, which contained 4.0 g dm−3 NaAlO2, 0.375 g dm−3 NaOH, 10.0 g dm−3 chromia, 0.8 g dm−3 Darvan C−N. Its conductivity was measured 5.3 ± 0.1 mS cm−1 and the pH value 11.1 ± 0.1. The compositions of electrolyte II were the same with those of electrolyte I, except the chromia pigment was GN-M, not self-made powder. The pH value of electrolyte II was 11.0 ± 0.1, and its conductivity was lower, 4.2 ± 0.1 mS cm−1. Several batches of self-made pigments were synthesized through precipitation and calcination. The precursor chemical, Cr(NO3)3· 9H2O (reagent grade, Choneye Chemicals), was dissolved in deionized water to make a 0.5 M solution, then precipitated with the solution of caustic soda (Fisher). The precipitates were washed with a sufficient amount of deionized water until the pH value of spent water reached 7, then calcined for 6 h at 450, 600, 650, 700 °C in air, and followed by pulverization with mortar and pestle. The calcined powder at 700 °C was chosen exclusively in preparing electrolyte I. The PEO treatment began with growth of a white barrier layer on the sample of 6061 Al alloy, followed by growth of the green layer. With the white barrier layer underneath, the flaws of green coating were easily identified through the naked eye. The surface area of the disk sample was 21.7 cm2. Pretreatment of the sample and the PEO setup had been described in the previous publication.37 The pulsed current was regulated by a high-voltage power supply (GX-100/1000, ADL GmbH), with different electrical settings for the two growth steps. In the barrier layer growth, the controlled voltage settings were denoted as 400 V(+)/100 V(−), since the square voltage waveform

3. RESULTS AND DISCUSSION 3.1. Pigment Characterization and Dispersion. Figure 1 presents the particle size distributions of pigment in electrolyte

Figure 1. DLS results of electrolyte I and II. The particle size of selfmade chromia powder is relatively small, 69 nm, with a narrow size distribution (PDI = 0.145). The average size of GN-M powder is 351 nm, and its distribution is narrower (PDI = 0.128). The samples subject to DLS analysis were the 1:500 dilutions of electrolyte I and II.

I and II, analyzed with DLS. The colloid of self-made pigment is indefinitely stable in electrolyte I. Its polydispersity index (PDI) is 0.145, considered to be monodisperse (