Printer Friendly

ON THE CUTTING PERFORMANCE OF DIAMOND-COATED CEMENTED CARBIDE TOOLS PRETREATED WITH MURAKAMI/[H.sub.2][O.sub.2]-[H.sub.2][SO.sub.4] SOLUTION.

ABSTRACT

In this study, substrate surface pretreatment using Murakami/[H.sub.2][O.sub.2]-[H.sub.2][SO.sub.4] solution was investigated as a method for improving the cutting performance of diamond-coated cemented carbide tools. Cutting performance of the coated tools with Murakami/[H.sub.2][O.sub.2]-[H.sub.2][SO.sub.4] and sand-blasting/HCI-[HNO.sub.3] pretreatments and an uncoated tool were examined in the continuous milling of medium density fiberboard, melapi (Shorea spp.), and western redcedar (Thuja plicata D. Don). Substrate surface pretreatment with Murakami/[H.sub.2][O.sub.2]-[H.sub.2][SO.sub.4] resulted in a considerable reduction of the initial edge dullness and edge roughness of the coated tool as compared to those obtained by utilizing the sand-blasting/HCI-[HNO.sub.3] pretreatment method. As a result, cutting-power consumption for the Murakami/[H.sub.2][O.sub.2]-[H.sub.2][SO.sub.4] tools was considerably lower than the sand-blasting/HCI-[HNO.sub.3] sandblasting pretreatment and at an approximately similar level as the uncoated tool, especially in the cutting of melapi. Delamination of the diamond film did not occur for any of the coated tools when cutting solid woods. In the case of cutting medium density fiberboard, the coated tools with Murakami/[H.sub.2][O.sub.2]-[H.sub.2][SO.sub.4] pretreatment exhibited far less delamination than the coated tools with sand-blasting/HCI-[HNO.sub.3] pretreatment.

The authors have made diamond-coated cemented carbide tools by the hot-filament chemical vapor deposition (CVD) method with a carbonized tantalum (TaC) filament [2], and the coated tools have been experimentally investigated for the possibility of their application to wood machining. In previous studies by the authors [3-9], the coated tools continuously milled four kinds of wood-based materials (cement-bonded particleboard, particleboard, medium density fiberboard (MDF), and plywood) and air-dried and wet solid woods (melapi and western redcedar). The results showed that the diamond-coated tools exhibited high resistance to mechanical and corrosive wear, especially in milling air-dried and wet solid woods.

The cutting-power consumption for the coated tools was higher than that for the uncoated tools for all work materials tested. This is believed to be caused by edge roundness due to coating thickness, the uneven surface of diamond film, and edge dullness attributed to sand-blasting pretreatment [3-9]. A previous study [7] showed that cutting-power consumption could be decreased by decreasing the diamond film thickness and by polishing the diamond film near the cutting edge. However, the level of cutting-power reduction brought about by these methods was not satisfactory. In addition, edge dullness and edge roughness brought about by sand-blasting may have detrimental effects on the quality of the machined surface.

In the study described in this paper, an alternative substrate surface pretreatment method was investigated. This pretreatment method involved chemical etching with Murakami ([K.sub.3] [Fe[(CN).sub.6]]: KOH:[H.sub.2]O = 1:1:10) and [H.sub.2][O.sub.2]-[H.sub.2][SO.sub.4] solutions [1] followed by ultrasonic scratching. Pretreatment with Murakami solution causes the formation of a rough surface for anchoring the diamond film. In addition, surface roughness levels caused by this treatment are considerably lower than those caused by sand-blasting. Cutting experiments were conducted with diamond-coated tools pretreated with Murakami/[H.sub.2][O.sub.2]-[H.sub.2][SO.sub.4] solution and with the sand-blasting method and the cuffing performances of the coated tools and an uncoated tool were compared.

EXPERIMENTAL CUTTING TOOLS

The cemented carbide insert in this study was similar to that in a previous study [8]. The insert was 17mm long and 12 mm wide and the substrate material was K05 grade (91% to 93% WC, 3.5% to 4.5% Co, and 3% to 5% other carbide). The sharpness angle was 55 degrees. In the cuffing experiments, tools were either coated (with two kinds of pretreatments) or uncoated.

Pretreatment of cemented carbide substrate. -- Two substrate surface pretreatment methods were utilized for enhancing the adhesion strength between the diamond film and the cemented carbide substrate. In the first pretreatment method, the cemented carbide inserts were dipped in Murakami solution for 30 minutes and then treated in 35 percent [H.sub.2][O.sub.2]-5vol.%[H.sub.2][SO.sub.4] solution for 10 seconds. For nucleation enhancement, the substrate surface was ultrasonically scratched in a solution of methyl alcohol and diamond particles (2 [micro]m to 3 [micro]m) for 30 minutes. The second pretreatment method included sand-blasting with #150 silcon carbide (SiC) particles, chemical etching of the cobalt from the surface with HCl-[HNO.sub.3] solution, and ultrasonic scratching with diamond particles. Details of this pretreatment method are provided elsewhere [3,5,8].

Scanning electron (SEM) micrographs and elemental analysis results of the cemented carbide substrate surfaces before and after pretreatment are shown in Figure 1. The elemental analyses were conducted by energy-dispersive x-ray spectroscopy (accelerating voltage, 15 kV). The substrate surface before pretreatment was smooth with visible grinding marks and cobalt smeared over the carbide grains (Fig. la). Large peaks for tungsten and small peaks for cobalt were exhibited by the elemental analysis of the untreated surface. After pretreatment with Murakami/[H.sub.2][O.sub.2]-[H.sub.2][SO.sub.4] solutions, the substrate surface exhibited a relatively uniform surface with fine pores caused by the selective etching of the tungsten carbide by the Murakami solution (Fig. 1b). Elemental analysis of this surface exhibited a peak for potassium (contained in the Murakami solution) in addition to peaks for tungsten and cobalt. The cobalt peak was shown to be slightly higher than that for the untreated tool. This may be at tributed to exposing the cobalt binder by the selective etching of tungsten carbide grains with Murakami solution [1]. The surface for the sandblasted and cobalt-etched substrate was remarkably uneven because of the sandblasting and its cobalt peak was evidently smaller due to the selective etching of cobalt by the HCI-[HNO.sub.3] solution.

Figure 2 shows SEM micrographs of the cutting edge as viewed from the side and in a direction perpendicular to the rake face for the tool before and after pretreatment. Before pretreatment, the tool cutting edge was sharp and straight. After pretreatment with Murakami/[H.sub.2][O.sub.2]-[H.sub.2][SO.sub.4], the cutting edge almost maintained the same level of sharpness and straightness as the untreated cutting edge. In contrast, pretreatment with sand-blasting/HCI-[HNO.sub.3] caused considerable roundness and jaggedness of the cutting edge. The figure shows that edge radius and edge roughness after Murakami/[H.sub.2][O.sub.2]-[H.sub.2][SO.sub.4] small as compared with those obtained by sand-blasting/HC1[HNO.sub.3] pretreatment.

Diamond coating.-- The diamond film was deposited from an H2-5vol.%[CH.sub.4] gas mixture (total pressure of 4 kPa) by the hot-filament CVD method (2). The filament power was 2.2 kW. The rake and clearance faces of the cemented carbide insert were coated with 5-, 10-, 15-, and 20-[micro]m-thick diamond films. SEM micrographs of the film surfaces, taken on the rake face of the coated tool with Murakami/[H.sub.2][O.sub.2]-[H.sub.2][SO.sub.4] pretreatment, are shown in Figure 3. Diamond crystals were found to grow larger with increasing film thickness. Figure 4 shows elemental analysis results of the 20-[micro]m-thick diamond films for both pretreatment methods. Similar results were obtained for other film thicknesses. A large cobalt peak is apparent in the elemental analysis of the diamond film deposited on the substrate pretreated with sand-blasting/HCl-[HNO.sub.3] This may be attributed to migration of cobalt and the formation of cobalt droplets on the film surface. In contrast, cobalt droplets were hardly o bserved on the film surface of the substrate pretreated with Murakami/[H.sub.2][O.sub.2]-[H.sub.2][SO.sub.4]. A possible explanation of this is the formation of thin films of the more stable cobalt compounds, CoS and Co[SO.sub.4] which prevent cobalt migration to the surface during diamond nucleation and growth [1].

SEM micrographs of cutting edge as viewed from the side and in a direction perpendicular to the rake face for the 20-[micro]m-coated tools are shown in Figure 5. In the case of the Murakami/[H.sub.2][O.sub.2]-[H.sub.2][SO.sub.4] pretreatment, edge radius was smaller and edge roughness was smoother compared to the sand-blasting/HCl-[HNO.sub.3] pretreatment. However, its edge radius was larger compared to the uncoated tool (Fig. 2a) because of the added film thickness. The edge radius of the coated tool became smaller with a decrease in the film thickness.

Figure 6 shows profiles of the straight cutting-edge of the uncoated and 20-[micro]m-coated tools. The method for recording the profiles was previously noted [9]. Table 1 lists cutting-edge roughness for each tool, represented by the maximum height, [R.sub.max] (Japanese Industrial Standards, JIS B 0601-1982). These values were calculated for each profile using the central 16-mm-long cutting edge. The cutting edge of the uncoated tool was the smoothest. The cutting edge of the coated tool with the Murakami/[H.sub.2][O.sub.2]-[H.sub.2][SO.sub.4] pretreatment was slightly rough. The roughness values for the various film thicknesses varied from 10.1 [micro]m to 11.3 [micro]m and therefore did not appear to depend on film thickness. The cutting edge of the coated tool with the sandblasting/HCl-[HNO.sub.3] pretreatment was remarkably rough. The cutting-edge roughness was two to three times larger than that for the coated tools pretreated with Murakami/[H.sub.2][O.sub.2]-[H.sub.2][SO.sub.4].

WORK MATERIALS

The work materials in this study included MDF as a wood-based material (specific gravity in air-dried condition, 0.67; moisture content, 9.2%) and air-dried melapi (Shorea spp. (Sabah); 0.48; 16.5%), and western redcedar (Thuja plicata D. Don; 0.33; 14.0%) as solid woods. All pieces of work material were 910 mm long, 300 mm wide, and 12 mm thick, and were machined on the 910-mm by 12-mm edge and in orthogonal 90 to 0 cutting for melapi and westem redcedar.

CUTTING TESTS

A series of cutting tests was carried out on a single-spindle shaper [8,9]. The materials were cut by up-milling. Each tool with a 15-degree clearance angle (rake angle, 20 degrees) was set on the milling cutter, which had a 130-mm-diameter cutting circle. The cutting conditions were: a spindle speed of 3800 rpm, a cutting speed of 25.9 in/second, a feed per revolution of 1.2 mm/rev, and a depth of cut of 2 mm. The total cutting length was approximately 226 m (total cutting arc length, approximately 3026 m; total net cutting time, approximately 50 min.).

MEASUREMENT METHOD

Edge recession and cutting-power consumption were measured and the cutting-edge profiles were recorded at prescribed cutting periods. SEM observations of the tool edge and calculations of the cutting-edge roughness were conducted for each tool at the end of the cutting test.

Edge recession was measured on the rake face with a microscope. A total of five equally spaced measurements were made along the 12-mm-long cutting edge and the edge recession was represented by the largest value. The edge recession for the coated tool was represented by the recession of the cemented carbide substrate [3-6,8,9]. The recession before cutting was approximately 11 [micro]m for the inserts pretreated by sand-blasting. In contrast, the initial recession for the Murakami/[H.sub.2][O.sub.2]-[H.sub.2][SO.sub.4] pretreated inserts was approximately 0 [micro]m.

Cutting-power consumption was measured with a digital three-phase wattmeter and a pen recorder. The cutting-power value was calculated by subtracting the power consumption during idling from that during cutting.

The methods for recording the cutting-edge profile and for calculating the cutting-edge roughness were as described previously.

RESULTS AND DISCUSSION

SEM OBSERVATIONS AND CUTTING-EDGE PROFILES

SEM micrographs of the 20-[micro]m-coated tools after cutting are shown in Figure 7, and the cutting-edge profiles of these coated tools are shown in Figure 8. Table 2 lists the cutting-edge roughness values for each tool after cutting. As shown in Figure 7a, the coated tools exhibited delamination of the diamond film during cutting MDF for both substrate surface pretreatment methods. For the coated tools with the Murakami/[H.sub.2][O.sub.2]-[H.sub.2][SO.sub.4] pretreatment, delamination occurred at a few points along the high density layer of the MDF edge for the 20 [micro]m thickness (Fig. 8a). However, delamination occurred along the entire cutting width, 12-mm-long cutting edge, for the other film thicknesses. In the case of the coated tools with the sandblasting/HCl-[HNO.sub.3] pretreatment, delamination occurred along the entire cutting width for all film thicknesses. Since the edge of each tool receded considerably, each roughness value was larger than the value before cutting as listed in Table 1.

In the case of cutting melapi and western redcedar, delamination of the diamond film was not visible and the cutting-edge profiles and cutting-edge roughness values did not change considerably during cutting. In contrast, the uncoated tools showed a larger roughness value than that before cutting. The cutting-edge roughness values for the coated tools with the Murakami/[H.sub.2][O.sub.2]-[H.sub.2][SO.sub.4] pretreatment were comparable to those of the uncoated tool.

EDGE RECESSION

The progression of edge recession with cutting distance for each cutting tool is shown in Figure 9. During the cutting of MDF, edge recession for the uncoated tool increased rapidly at the initial stages of cutting and then progressed gradually. Edge recession for the 5-[micro]m-, 10-[micro]m-, and 15-[micro]m-coated tools with the Murakami/[H.sub.2][O.sub.2]-[H.sub.2][SO.sub.4] pretreatment also increased rapidly at the initial stages of cutting, which is attributed to diamond film delamination. At the end of the cutting test, these tools attained similar levels of edge recession compared to the uncoated tool. In the case of the 20-[micro]m-coated tool, delamination occurred at a later stage (cutting length, approx. l00m) and thus the final edge recession for this cutting tool was considerably smaller than that for the uncoated tool. Similarly, edge recession for the 5-[micro]m-, 10-[micro]m-, and 15-[micro]m-coated tools with the sand-blasting/HCl-[HNO.sub.3] pretreatment progressed rapidly at the early st ages of cutting and attained similar levels of wear compared to the uncoated tool at the end of the cutting test. Delamination for the 20-[micro]m-coated tool also occurred at an early stage (cutting length, approx. 20 m) but its final edge recession value was smaller than that for the uncoated tool.

In the cutting of melapi and western redcedar, the edge recession for the uncoated tools increased at a slower rate than that for cutting MDF. Edge recession at the final cutting period was slightly higher when cutting melapi than that when cutting redcedar. Edge recession for the coated tools did not change from its initial value throughout the cutting period since delamination of the diamond film was not evident for any of the coated tools. The edge recession for each coated tool at the final cutting period was smaller than for the uncoated tool.

CUTTING-POWER CONSUMPTION

Variations in cutting-power consumption for each cutting tool over the whole cutting period are shown in Figure 10. The power consumption value for the coated tools tended to increase with an increase in film thickness for each work material.

During cutting of MDF, the power consumption for the uncoated tool increased rapidly until the cutting length reached approximately 50 m and then increased gradually. This tendency almost corresponded to the result in the edge recession. The power consumption value for the coated tools also showed a similar tendency compared to the uncoated tool and tended to increase with the occurrence of delamination. The coated tools with the Murakami/[H.sub.2][O.sub.2]-[H.sub.2][SO.sub.4] pretreatment showed higher cutting-power consumption values than the uncoated tool. However, except for the 15-[micro]m-coated tool, the cutting-power consumption values after a cutting length of approximately 50 m were approximately the same as the uncoated tool. The coated tools with the sand-blasting/HC1-[HNO.sub.3] pretreatment consistently showed larger cutting-power consumption values compared to the uncoated tool. This is attributed to edge dullness due to both film thickness and sand-blasting.

In the case of cutting melapi, the cutting-power consumption for the uncoated tool hardly increased over time, compared to the increase that took place during the cutting of MDF. The coated tools with the sand-blasting(HC1-[HNO.sub.3] pretreatment consistently showed a larger cutting-power consumption value compared to the uncoated tool. In contrast, the differences in the values of cutting-power consumption between the coated tools with the Murakami/[H.sub.2][O.sub.2]-[H.sub.2][SO.sub.4] pretreatment and the uncoated tool were considerably less, especially in the case of the 5-[micro]m-coated tool.

In the cutting of western redcedar, the results were almost the same as in the cutting of melapi. However, the remarkable effect of Murakami/[H.sub.2][O.sub.2]-[H.sub.2][SO.sub.4] pretreatment on cutting-power consumption for the coated tools was not as apparent as in the case of cutting melapi.

SUMMARY

In this study, substrate surface pretreatment using Murakami/[H.sub.2][O.sub.2]-[H.sub.2][SO.sub.4] solution was investigated as a method for improving the cutting performance of diamond-coated cemented carbide tools. Cutting performance of the coated tools with the Murakami/[H.sub.2][O.sub.2]-[H.sub.2][SO.sub.4] and sand-blasting/HCI-[HNO.sub.3] pretreatments and the uncoated tools was examined in the continuous milling of MDF, melapi, and western redcedar. The results obtained are summarized as follows:

1. The edge dullness and edge roughness of the coated tools with the Murakami/[H.sub.2][O.sub.2]-[H.sub.2][SO.sub.4] pretreatment were considerably improved compared to the coated tool with the sand-blasting/HC1[HNO.sub.3] pretreatment.

2. During the cutting of MDF, delamination occurred for each film thickness for both pretreatment methods. For the 20-[micro]m film thickness, delamination for the coated tool pretreated with Murakami/[H.sub.2][O.sub.2]-[H.sub.2][SO.sub.4] occurred at a much later stage compared to the tool pretreated with sand-blasting/HC1-[HNO.sub.3]. This may be attributed to enhancements of the interface between the diamond film and the substrate as caused by the Murakami/[H.sub.2][O.sub.2]-[H.sub.2][SO.sub.4] pretreatment.

3. In the case of cutting of melapi and western redcedar, delamination did not occur at all and the edge recession did not increase for any of the coated tools. The coated tools with the Murakami/[H.sub.2][O.sub.2]-[H.sub.2][SO.sub.4] pretreatment exhibited a decrease in the cutting-power consumption, especially in the cutting of melapi. The 5-[micro]m-coated tool with the Murakami/[H.sub.2][O.sub.2]-[H.sub.2][SO.sub.4] pretreatment consistently showed an approximately similar cutting-power consumption level compared to the uncoated tool.

The authors are, respectively, Research Student, Faculty of Agriculture, Kyushu Univ., Hakozaki 6-10-1, Higashi-ku, Fukuoka 812-8581, Japan; Assistant Professor, Dept. of Industrial and Manufacturing Engineering, Wichita State Univ., Wichita, KS 67260-0035; Professor, Faculty of School Education, Hiroshima Univ., Kagamiyama 1-1-1, Higashi-Hiroshima 739-8524, Japan; Researcher, Hiroshima Prefectural Western Industrial Research Inst., Agaminami 2-10-1, Kure 737-0004, Japan; and Professor, Faculty of Agriculture, Kyushu Univ. The results of this work were presented at the 49th Ann. Meeting of the Japan Wood Res. Soc., April 2-4, 1999, Tokyo, Japan. The authors would like to thank Mr. Masanobu Kamata and researchers of Hiroshima Prefectural Western Industrial Res. Inst. for kind cooperation in making diamond-coated cemented carbide tools. This work was partly supported by the Sasakawa Scientific Res. Grant (No.10-178) from the Japan Sci. Soc. This paper was received for publication in April 1999. Reprint No. 897 1.

LITERATURE CITED

(1.) Haubner, R., S. Kubelka, B. Lux, M. Griesser, and M. Grasserbauner. 1995. Murakami and [H.sub.2][SO.sub.4]/[H.sub.2][O.sub.2] pretreatment of WC-Co hard metal substrates to increase the adhesion of CVD diamond coatings. J. de Physique IV C5:753-760.

(2.) Matsubara, H. and T. Sakuma. 1990. Diamond deposition on cemented carbide by chemical vapor deposition using a tantalum filament. J. of Materials Sci. 25:4472-4476.

(3.) Morita, T., K. Banshoya, T. Tsutsumoto, and Y. Murase. 1995. Cutting of difficult-to-cut wood-based materials with diamond-coated cemented carbide tools. Part 1. Difference of cutting performance by the filament power in the synthesis of diamond film. Mokuzai Gakkaishi 41(12):1093-1101. (in Japanese, English abstract).

(4.) _____, _____, _____, and _____. 1996. Application of diamond films to cutting tools for difficult-to-cut wood-based materials. New Diamond 12(3):26-27. (in Japanese).

(5.) _____, _____, _____, and _____. 1995. Cutting performance of diamond-coated cemented carbide tools. In: Proc. of the 12th Inter. Wood Machining Seminar, Kyoto, Japan. pp. 302-313.

(6.) _____, _____, _____, and _____. 1997. Cutting of difficult-to-cut wood-based materials with diamond-coated cemented carbide tools. Part 2. Difference of cutting performance by sharpness angle. Wood Industry 52(4):194-198. (in Japanese, English abstract).

(7.) _____, _____, _____, M. Kawamitsu, and Y. Murase. 1997. Characteristics of diamond-coated cemented carbide tools in the milling of particleboard. In: Proc. of the 13th Inter. Wood Machining Seminar, Vancouver, Canada. pp. 687-698.

(8.) _____, _____, _____, and Y. Murase. 1998. Effects of work materials on cutting performance of diamond-coated cemented carbide tools. Forest Prod. J. 48(5):43-50.

(9.) _____, _____, _____, and _____. 1999. Corrosive-wear characteristics of diamond-coated cemented carbide tools. J. of Wood Sci. 45(6):463-469.
 Cutting-edge roughness ([R.sub.max],
 [micro]m) for coated and uncoated tools used in the study.
 Murakami/[H.sub.2][O.sub.2]-
 [H.sub.2][SO.sub.4]
Uncoated 5 [micro]m 10 [micro]m 15 [micro]m 20 [micro]m
 7.4 10.1 11.0 11.3 11.0
 Sand-blasting/
 HCl-[HNO.sub.3]
Uncoated 5 [micro]m 10 [micro]m 15 [micro]m 20 [micro]m
 7.4 30.5 31.0 27.4 25.2
 Cutting-edge roughness ([R.sub.max],
 [micro]m) for coated and uncoated tools
 after the cutting tests.
 Murakami/[H.sub.2][O.sub.2]
 [H.sub.2][SO.sub.4]
Work materials Uncoated 5 [micro]m 10 [micro]m 15 [micro]m
MDF 44.4 60.3 58.8 60.2
Melapi 10.1 10.7 12.4 13.3
Western redcedar 13.5 10.2 11.3 11.2
 Sand-blasting/
 HCl-[HNO.sub.3]
Work materials 20 [micro]m 5 [micro]m 10 [micro]m 15 [micro]m
MDF 54.3 36.2 46.4 48.8
Melapi 10.9 28.5 28.5 26.6
Western redcedar 11.0 32.1 31.7 29.3
Work materials 20 [micro]m
MDF 47.0
Melapi 29.2
Western redcedar 24.3
COPYRIGHT 2000 Forest Products Society
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2000 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:MORITA, TAKAO; SHEIKH-AHMAD, JAMAL Y.; BANSHOYA, KAORU; TSUTSUMOTO, TAKAHIRO; MURASE, YASUHIDE
Publication:Forest Products Journal
Date:Jan 1, 2000
Words:3892
Previous Article:CANADIAN BLUESTAIN FUNGI: VARIATION IN TOLERANCE TO SAPSTAIN CONTROL BIOCIDES.
Next Article:COMPARISON OF ACTUAL VS. SIMULATED FAILURE DISTRIBUTIONS OF FLEXURAL WOOD SPECIMENS SUBJECT TO 5-DAY LOAD SEQUENCES.

Terms of use | Privacy policy | Copyright © 2021 Farlex, Inc. | Feedback | For webmasters |