Influence of the substrate hardness on the cavitation erosion ...

Wear 263 (2007) 395–401

Influence of the substrate hardness on the cavitation erosion resistance of TiN coating Alicja Krella a,∗ , Andrzej Czyzniewski b a

b

The Szewalski Institute of Fluid-Flow Machinery, Polish Academy of Sciences, ul. Fiszera 14, 80-231 Gdansk, Poland Technical University of Koszalin, Department of Materials Science and Engineering, Raclawicka 15-17, 75-620 Koszalin, Poland Received 5 August 2006; received in revised form 14 December 2006; accepted 1 January 2007 Available online 23 May 2007

Abstract Nanocrystalline TiN coating was deposited by means of the cathodic arc method on stainless steels types X6CrNiTi18-10 and X39Cr13. Both steels were subjected to thermal treatment in order to obtain substrates of different hardness: 1.7 GPa, 2.8 GPa and 4.6 GPa. The TiN coating was 3.7 ␮m thick. The TiN coating has strong (1 1 1) crystallographic orientation and fine crystalline structure of ␦-TiN phase. The TiN coating is characterized by high hardness (25.4 GPa) and good adhesion. The adhesion increases with the substrate hardness. The evaluation of TiN coating resistance to cavitation erosion is based on the investigation performed in a cavitation tunnel with a slot cavitator and tap water as a medium. The estimated cavitation resistance parameters of coating were the incubation period of damage and the total mass loss after the whole test. It has been confirmed that the incubation periods of the coated steels were from 2 to 4 times longer than that of the uncoated steels. The mass losses of the coated steels decrease approximately 2.5 times in comparison with the uncoated steels. The scanning microscope analysis indicates that the damage of TiN coating is mainly due to its delamination. The character of the coating and substrate damage in multiple locations indicates that the hard coating micro-particles torn off during the cavitation bubbles implosion hit against the coating and the revealed areas of substrate. As a result, the coating and especially the substrate of relatively low hardness beside cavitation erosion are subject to solid particle erosion with the hard torn off micro-particles of coating. © 2007 Elsevier B.V. All rights reserved. Keywords: Nanocrystalline TiN coating; Stainless steel; Cavitation erosion

1. Introduction TiN coatings are widely used for the protection of tools. These coatings generally provide high hardness, low coefficient of friction, good corrosion and oxidation wear resistance [1,2]. TiN coatings possess good impact resistance and cavitation resistance [1,2]. Cavitation phenomena usually occur in fluid systems where there are strong pressure fluctuations in fluids [3]. These fluctuations nucleate bubbles, which implode causing highenergy impact on solid surface. These implosions can fracture surface coating and can dislodging particles from the surface. Generally fine-grained materials with high Young modulus, high hardness, and smooth surface are best for resisting cavitation damage [4–7]. The mechanism of cavitation resistance of coated materials is more complicated due to the mismatch of mechan∗

Corresponding author. Tel.: +48 58 341 12 71; fax: +48 58 341 61 44. E-mail address: [email protected] (A. Krella).

0043-1648/$ – see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.wear.2007.02.003

ical properties and the adhesion of the coating to the substrate [8]. This paper reports the results of a study of the cavitation resistance of 3.7 ␮m thick TiN coatings on three hardnesses of stainless steel. Authors expect that higher substrate hardness causes better hard coating protection in cavitation erosion as a result of less plastic flow and therefore better adhesion. 2. Experimental details 2.1. Materials Stainless steel X39Cr13 and X6CrNiTi18-10 were used as the substrate materials. The chemical compositions of steels are presented in Table 1. The specimens made of the X6CrNiTi1810 steel were subjected to solutioning at 1050 ◦ C. The obtained hardness was 1.7 GPa. The specimens made of the X39Cr13 steel were subjected to quenching at 1050 ◦ C and tempering at 400 ◦ C and 600 ◦ C to obtain hardness of 4.6 GPa and 2.8 GPa, respec-

396

A. Krella, A. Czyzniewski / Wear 263 (2007) 395–401

Table 1 The chemical composition of steels Content of an element [%]

X39Cr13 X6CrNiTi18-10

C

Mn

Si

Ni

Ti

Cr

P

S

0.44 0.014

0.5 1.65

0.47 0.61

– 9.24

– 0.23

13.31 17.36

0.025 0.024

0.030 0.029

tively. All specimens (45 mm × 26 mm × 14 mm) were ground with a series of emery papers of 120, 320 and 600 grits, the final step was performed with diamond grinding paste to achieve the roughness of Ra ≤ 0.05 ␮m. The uncoated specimens were marked 1.7, 2.8 and 4.6, where the number means the hardness of the specimen. 2.2. Deposition and investigation of the TiN coating The TiN coating was deposited by cathodic arc method (ARC) in a vacuum chamber equipped with arc sources, with a target 100 mm in diameter. The 99.9% pure titanium target, 99.995% pure argon and 99.995% pure nitrogen gases were applied. Substrates preparation for the process consisted of ultrasonic-aided cleaning with organic solvents and alkaline detergents. The typical process of coating deposition includes the following operations: pumping-off air from the chamber down to the pressure below 2 × 10−3 Pa; heating the substrates to the temperature of 350 ◦ C; cleaning of the substrates by argon and titanium ions; deposition of thin titanium interlayer (∼0.05 ␮m thickness) in the argon atmosphere; deposition of TiN coating in the nitrogen atmosphere up to the thickness of approximately 3.7 ␮m. Parameters applied to the deposition of TiN coating are presented in Table 2. The coated specimens were marked TiN-1.7, TiN-2.8 and TiN-4.6, where 1.7 means X6CrNiTi18-10 steel with hardness of 1.7 GPa, 2.8–X39Cr13 steel with hardness of 2.8 GPa, and 4.6–X39Cr13 steel with hardness of 4.6 GPa. Basic properties of TiN coating deposited on X39Cr13 and on X6CrNiTi18-10 steels are shown in Table 3. The phase composition of the coating measured on a DRON2 X-ray diffractometer using Co K␣ radiation showed that the TiN coating is consistent with phase of ␦-TiN, what was confirmed by strong (1 1 1) plane reflection (Fig. 1). The grains size was determined by means of the Scherrer method with reflex parameters (location and FWHM) using Gaussian analysis. The size of TiN crystallites was estimated to be approximately 16 nm. The coating morphology and thickness were examined with JEOL JSM 5500 LV scanning electron microscope (SEM). The microphotograph

of the cross-section and the surface of TiN coating are shown on Fig. 2. Impurities visible on the coating surface are the microdroplets of titanium. These microdroplets always occur when the cathodic arc plasma method is used. The hardness and Young modulus were measured with a NanoHardness Tester (CSEM) using the method of Oliver and Pharr [9]. The maximum indentation depth of 300 nm was applied. The obtained hardnesses and Young modulus are presented in Table 3. A scratch tester Revetest® produced by CSEM was used to investigate the adhesion of coating. A diamond indenter with radius 0.2 mm was used and measurements were carried out at normal loading rate of 100 N/min, scratch speed of 10 mm/min and scratch length of 10 mm. At least three scratches were done. The first minor cracks for TiN-4.6, TiN-2.8 and TiN-1.7 occurred with a scratch made under the load of 29N, 20 N and 10N, respectively (Fig. 3). These loads correspond to the critical load LC1 defined by the occurrence of the first cohesive failure of the Table 3 Properties of TiN coating

Phase composition, structure Crystallographic orientation Mean crystallites size (nm0 Hardness (GPa) Young’s modulus (GPa) Coating thickness (␮m) Adhesion LC1 (N) LC2 (N) Roughness Ra (␮m)

TiN-4.6

TiN-2.8

TiN-1.7

␦-TiN (1 1 1) 16 ± 2 25.4 ± 2.5 295 ± 38 3.7 ± 0.1

␦-TiN (1 1 1) 16 ± 2 25.4 ± 2.5 295 ± 38 3.7 ± 0.1

␦-TiN (1 1 1) 16 ± 2 25.4 ± 2.5 295 ± 38 3.7 ± 0.1

29 ± 3 48 ± 3 0.31 ± 0.03

20 ± 3 36 ± 3 0.35 ± 0.03

10 ± 3 23 ± 3 0.35 ± 0.03

Table 2 Deposition parameters of TiN coating Pressure of residual gases Working pressure of argon Working pressure of nitrogen Arc current Substrate bias voltage Substrate temperature Target–substrate distance

2 × 10−3 Pa 1 Pa 1 Pa 80 A −100 V ∼350 ◦ C 150 mm Fig. 1. X-ray diffraction patterns of the TiN coating.


Fig. 2. The cross-section and the surface of the TiN coating, SEM.

coating, that are the cracks caused by tensile stresses inside and on the edges of the scratch behind the sliding diamond cone. The critical loads LC2, at which the coating removal from inside of the scratch starts, were 48 N, 36 N and 23 N, respectively. The performed examinations show that the hardness of substrate has an influence on coating adhesion. Along with the increase of substrate hardness the increase of adhesion occurred. The stylus profilometer Hommel-Tester 2000 was used to measure roughness (Ra ) the substrate and TiN coating. 2.3. Procedure of cavitation erosion test The experimental tests were performed in a cavitation tunnel equipped with a system of barricades. The schematic of the cavitation chamber is shown in Fig. 4. Cavitation intensity is controlled by adjusting the slot witdh and the boost pomp speed. Flow conditions are defined by the p1 and p2 absolute pressures measured at the chamber inlet and outlet, respectively. Vertical positioning of the specimen is easily adjustable by means of spacer washer. Tap water is used as the working liquid. The cavities (cavitating vortices and bubbles) are generated by the pressure decrease in the slot between two semi-cylindrical barricades. The TiN coated and uncoated specimens were subjected to cavitation impingement in the cavitation chamber operated with inlet pressure p1 = 1000 kPa, outlet pressure p2 = 130 kPa, and

397

Fig. 4. Schematic view of cavitation chamber with a system of barricades. I: stationary barricade; II: moving counter-barricade.

the slot width – 5 mm. Tests were performed without spacer washer. In order to obtain the erosion curves, the mass loss was measured after each exposure interval. The specimens were cleaned, dried and weighed before the test and after each test interval. Mass losses of the tested specimens were measured using an analytical balance with the permissible error limit of the balance ±1.4 mg for the load up to 100 g. At the beginning of the cavitation test the measurements were conducted every 30 min of exposure (for the first 180 min of test) to estimate the incubation period, and then the duration of exposure intervals was gradually increased. The total cavitation test exposure was 600 min. After each particular exposure time the cavitation erosion damage was analysed with macroscopic sample surface observation and the microstructure was investigated using the scanning electron microscope. 3. Results The mass losses of TiN coated and uncoated specimens arising during cavitation tests are shown in the graphic form in Fig. 5. The incubation period of all uncoated specimens was identical, and lasted less than 90 min. The least mass loss occurred on X6CrNiTi18-10 austenitic steel (7 mg), while the highest mass loss occurred on X39Cr13 steel with hardness of

Fig. 3. TiN coating damage arising during scratch test.

398


Fig. 5. Erosion curves of uncoated and TiN coated specimens.

4.6 GPa (11 mg). The SEM images of damaged specimens after the whole test are shown in Fig. 6. The surface of uncoated X6CrNiTi18-10 steel under cyclic action of cavitation pulses underwent undulation; originated slip bands agglomerates and big twins (Fig. 6a). The microcracks have been initiated at intrusions in slip bands (Fig. 6a) like in fatigue [10]. Microstructure of uncoated X39Cr13 (both hardness) steel under action of cavitation has been disclosed (Fig. 6 b, c). In case of X39Cr13 steel with hardness of 2.8 GPa, degradation has proceeded along martensite lathings causing material groove (Fig. 6b), therefore the shape of martensite lathings are not so clear as that arisen on X39Cr13 steel with hardness of 4.6 GPa (Fig. 6c). On X39Cr13 steel with hardness of 2.8 GPa some pits/microtunels (arrows) are also observed in Fig. 6b. In case of X39Cr13 steel with hardness of 4.6 GPa, because of high hardness the deformation is restrained in martensite lathings. Material is crumbled along streak lines (Fig. 6c). These lines are combinated with water flow and construction of cavitation tunnel. The best cavitation erosion resistance of uncoated austenitic steel is probably linked with absorption some degradation energy on plastic deformation (the surface undulation and the origin of some slip bands – Fig. 6a) and phase transformation due to cyclic impact during cavitation test (Fig. 7). The X-ray diffraction pattern of the X6CrNiTi18-10 steel after the cavitation test (Fig. 7) shows the increase of diffraction peak of Fe-␣ (1 1 1) and additionally appearance of the Fe-␣ (2 0 0), Fe-␣ (2 1 1) and Fe-␣ (2 2 0) reflections and the weakness of Fe-␥ reflections. The temporary arrest of mass loss that occurred between 90 and 120 min of erosion was most likely correlated with the phase transformation. The phase transformation, in turn, caused changes in the mechanical and physical properties on the specimen surface. Deposition of TiN coating on specimens caused the lengthening of incubation period and the decrease of mass loss for all tested coated specimens in comparison to uncoated specimens after the whole cavitation test (Fig. 5). The incubation period amounted