Coated cutting tool insert and fabrication method

(54) Title: CVD coated carbide, cermet or ceramic inserts and methods of manufacturing the same (56) Published publications: - (47) Abstract:

The present invention relates to a CVD coated notch with improved toughness properties having a TiC_xNew layers having a low tensile stress level of 10-300 MPa and a (X-ÄI2O3 layer with a high surface smoothness of <0.1 µm as measured by ABM technology) are obtained by subjecting the coating to a first intensive wet blasting operation with a slurry consisting of F80 grains of AI2O3 followed by a second wet blasting treatment with a slurry of F320 grains of AI2O3.

528 696

Summary

The present invention relates to a CVD coated insert with improved toughness properties having a TiC_xNew layers having a low tensile stress level of 10-300 MPa and a (X-AI consisting of F80 grains of AI2O3 followed by a second wet blasting treatment with a slurry of F320 grains of AI2O3.

12087 swe_ny11 summary.doc

528 696

The present invention relates to a CVD coated cutter suitable for machining metals with turning, milling, drilling or similar chip forming machining methods. The coated insert shows improved toughness behavior when using intermittent machining operations.

Modern, highly productive, chip-forming machining of metals requires reliable inserts with high abrasion resistance, good toughness properties and excellent resistance to plastic deformation.

This has so far been achieved by the use of a cemented carbide insert coated with a durable coating. The cemented carbide insert is usually in the form of a tapered insert in a tool holder, but can also be in the form of a solid cemented carbide drill or a cutter. Carbide inserts coated with various types of hard layers such as TiC,

TiC_xNew, TiN, TiC_xNYO_z and AI2O3 have been commercially available for many years. Such coatings are usually made up of several hard layers in a multilayer structure. The order and thickness of the individual layers are carefully selected to suit different application areas and work materials.

The coatings are usually precipitated by chemical vapor deposition (CVD) or physical vapor deposition (PVD) techniques. In some rare cases, plasma assisted chemical vapor deposition (PACVD) has also used s.

CVD technology is often preferred over PVD, as it has several advantages. It allows larger coating batches, produces coatings with good thickness distribution on complex shaped inserts, has a high coating rate, can be used to deposit electrically non-conductive layers such as Al₂O3 and ZrO₂. Many different materials can be deposited in the same coating run such as e.g. al₂O₃, TiC, TiC_xNew, TiN, TiC_xNYO_z, ZrC_xN_y and ZrO₂.

The CVD technology is performed in a fairly high temperature range, 950-1050 ° C. Due to this high coating temperature and a difference in thermal expansion coefficient between the deposited coating materials and the cemented carbide, CVD produces coatings with cooling cracks and tensile stresses.

PVD processes are run at a significantly lower temperature, 450-650 ° C and are carried out under heavy ion bombardment, resulting in crack-free layers with high compressive stresses. The high compressive stresses and the absence of cooling cracks make PVD-coated inserts much tougher 12087swe_ny description.doc

528 696 than CVD coated inserts and is therefore often preferred in intermittent cutting operations such as milling.

A noticeable improvement in the performance of CVD coated inserts came when MTCVD (medium temperature CVD) technology began to enter the tool industry about 5-10 years ago. An improvement of the cutting toughness properties was obtained. Today, a majority of tool manufacturers use this technology. Unfortunately, the MTCVD technology is limited only to the production of TiC_xNew layers of about 0.5 <x <0.7 and 0.3 <y <0.5 and x + y equal to or close to 1. The coating process here takes place at temperatures in the range 700-930 ° C. It uses a gas mixture of CH 3 CN, T 1 Cl 4 and Hg. Today's modern coatings also include at least one layer of AI2O3 to provide high pit wear resistance.

A further improvement in toughness properties could be obtained when coating temperature for the 01-AI2O3 process was also possible as shown in patent application EP-A-1464727.

Finishing of coated inserts by brushing or by wet blasting is shown in numerous patents. The intention is to provide a smooth cutting edge and / or to expose AI2O3 along the edge line, e.g. shown in US 5,851,687 and in EP 603 144 or to obtain AI2O3 as the top layer even on the chip side in cases where TiN is used as a wear detection layer on the release side as shown in US 5,861,210. Any treatment technique that exposes a surface such as e.g. a coating surface, for a strong pulse such as e.g. wet or dry blasting or ultrasonic waves will have an effect on the coating stress state (σ). However, to significantly reduce the tensile stresses in all layers of a CVD coating structure, intensive surface treatment is required. However, such treatment can even lead to an excessive change in the voltage state, e.g. from a high tensile to high compressive stress as shown in EP-A-1311712, in which a dry blasting technique is used.

For the wet blasting technique, the blasting medium, usually Al2O3 grains and water, must hit the coating surface with a high impulse. The impulse can be controlled by e.g. the pressure of the blasting pulp, the distance between the blasting nozzle and the coating surface, the grain size of the blasting medium, the concentration of the blasting medium and the angle of impact of the blasting jet.

Despite these advances, further improvements in toughness properties for CVD coated inserts are highly desirable.

12087swe_ny description.doc

528 696

It is an object of the present invention to provide CVD coated tool inserts with improved toughness properties.

It is a further object of the present invention to provide a method of manufacturing CVD coated tool inserts with improved toughness properties.

Fig. 1 shows a goniometer set for determining residual stress with X-ray measurements in which

E = Euler% cradle

S = sample

I = incoming X-ray

D = diffracted X-ray

Θ = diffraction angle ω = θ ψ = angle of inclination along the Euler K cradle

Φ = rotation angle around the sample shaft

The present invention therefore relates to coated inserts comprising a body of generally polygonal or round shape having at least one chip side and at least one release side comprising a coating and a substrate. The coating comprises at least one TiC_xNew layer and a well-crystalline layer consisting of 100% α-Αΐ₂03. Such an α-Al₂O3 ~ layer is the visible top layer at least on the chip side and along the edge line and it has been wet blasted intensively or treated with some other similar technique capable of bombarding the coating surface with a sufficiently high energy to create tensile stress relaxation in both AI2O3 and TiC_xNew layers. The AI2O3 top layer has a very smooth surface.

Surprisingly, it has been found that a significantly improved toughness performance can be obtained if a coated cutting insert with a generally polygonal or round shape having at least one chip side and at least one release side where the cutting insert is at least partially coated has the following characteristics:

- a second-tier TiC_xNew layers having a thickness of 1-8 µm, preferably 2-5 µm, where x> 0, y> 0 and x + y = 1, preferably prepared with MTCVD, with tensile stresses 10-300 MPa, preferably 10-200 MPa. and

- an outer α-Al 2 C> 3 layer having a thickness of 1-5 µm, preferably 2-4 µm, being the top layer on the chip side and along the edge line with a mean surface fineness Ra <0.1 µm over a length of 10 µm with Atomic Force Microscopy (AFM) and a relationship in

12087swe_ny description.doc

528 696 X-ray diffraction intensity (peak height minus background) of

1 (012) / I (024)> 1.5.

Preferably there is a bonding layer of TiC_xNYO_z, x> 0, z> 0 and y> 0 between TiC_xThe new layer and the a-Al2C> 3 layer. The total thickness of the two layers is <10 µm, preferably <6 µm.

Further layers may be incorporated into the coating structure between the substrate and the layers of the present invention consisting of metal nitrides and / or metal carbides and / or metal oxides with the metal element selected from Ti, Nb, Hf, V, Ta, Mo, Zr, Cr, W and Al to a total coating thickness of <20 µm.

It is preferred to have low tensile stresses in TiC_xThe new layer because it was found that if compressive stresses were induced by biasing, very high impulse was required for the blasting and under such conditions, peeling of the coating occurred along the cutting edge. It was also found that such induced compressive stresses were not stable with respect to temperature increases occurring in a cutting operation as compared to whether the coating retains some compressive stress.

The remaining voltage, σ, of the inner TiC_xThe new layer is determined by XRD measurements using the well-known sin²The method as described by IC Noyan, JB Cohen, Residual Stress Measurement by Diffraction and Interpretation, Springer-Verlag, New York, 1987 (pp. 117-130). The measurements should be carried out using CuKa radiation on TiC_xThe new (422) reflex with a goniometer set as shown in Fig. 1. The measurements should be performed on as flat a surface as possible. It is recommended to use ψ geometry with six to eleven ψ angles, equidistant within one²i µ range of 0 to 0.5 (ψ = 45 °). An equidistant distribution of Φ angles within a Φ sector of 90 ° is also preferred. To verify a biaxial stress state, the sample should be rotated for Φ = 0 ° and 90 ° simultaneously inclined in ψ. It is recommended to investigate the possible presence of shear stresses and therefore both negative and positive ψ angles should be measured. In the case of an Euler 1/4 cradle, this is achieved by measuring the sample even at Φ = 180 ° and 270 ° for the different ψ angles. Its²The method is used to determine the residual voltage preferably using any commercially available software such as DIFFRAC^plus Stress32 v. 1.04 from Bruker AXS with the constant Emodule, E = 480 GPa and Poisson's ratio, v = 0.20 in case of an MTCVD Ti (C, N) layer and localization of the reflex using Pseudo-Voigt-Fit -funtionen. In that case, the following parame12087swe_ny description.doc is used

528 696 steps: E-module = 480 GPa and Poisson's ratio v = 0.20. In the case of a biaxial stress state, the tensile stress is calculated as the average of the biaxial stresses obtained.

For a-Al2C> 3, it is usually not possible to use sin²The technique because the necessary high 20-angle reflexes are often too weak. However, the inventor has found a useful alternative measurement that relates the state of OI-AI2O3 to cutting performance.

For an α-Al 2 C> 3 powder, the ratio of the diffraction intensities 1 (012) / 1 (024) is close to 1.5. Powder diffraction card JCPDS No. 43-1484 shows the intensities I_O(012) = 72 and I_O(O24) = 48th

The inventor has observed that for CVD-C1-AI2O3 layers on cemented carbide with tensile stress (σ about> 350 MPa) the intensity ratio 1 (012) / 1 (024) is significantly lower than the expected value 1.5, usually <1. This can be is due to some disorder in the crystal lattice caused by the tensile stresses. It has been found that when such a layer is subjected to intense blasting (free from stresses) or if it is completely removed from the substrate and pulverized (not under tension), the ratio 1 (012) / 1 (024) becomes closer, equal or even higher. than 1.5 depending on the blasting power used. Thus, this intensity ratio can be used as an important state property for a CX-AI2O3 layer. A ratio higher than 1.5 may occur if a high blasting force is used in conjunction with the fact that for thin layer XRD analyzes, the low 20 angle such as the (012) reflex peak intensity 1 (012) will be too highly estimated in comparison. with higher angular reflexes e.g. (024) reflex unless a so-called thin film correction calculation is performed.

According to the method of the present invention, a cutting insert is provided with a CVD coating comprising a second outer TiC_xNew layer and outer CI-AI2O3 layer. To obtain a high surface smoothness and low tensile stress level, the coating is subjected to a first intense wet blasting operation on the chip side with a slurry consisting of F80 grains (FEPA standard) of AI2O3 in water at an air pressure of 1.8-2.4 bar for about 2 hours. -8 sec / cut followed by a second blasting treatment with a slurry of F320 grain (FEPA standard) of AI2O3 in water with an air pressure of about 2 bar for about 4-10 sec / cut. A freshly mixed slurry of F80 barley and water would first be used on dummy inserts before use on production cutters to round the barley slightly.

12087swe_ny description.doc

528 696

If a smooth coating surface is also desired on the release side, blasting can also be performed there. In this case comes TiC_xNew layer to have low tensile stresses both on the chip side and on the release side. This is the preferred embodiment of machining forging materials such as stainless steel or nodular cast iron.

If cutting is desired with a different color on the release side than on the black chip side, a thin outer 0.1-2 μιη color layer of TiN (yellow), TiC is deposited._xNew (gray or bronze) or TiC (gray). The inserts are then blasted at a 90 ° spray angle which removes the top layer and exposes the black Al₂O₃layer. In this case, the coating on the chip side will have the low desired tensile stress, while the release side will have high tensile stresses in the range of 600-1000 MPa depending on the choice of coating and the length extension coefficient (CTE) of the cemented carbide insert used.

Example 1

A) R390-11T308M-PM cemented carbide inserts with the composition 12.6 wt% Co, 1.25 wt% TaC, 0-32 wt% NbC and residual WC (CTE = about 6 * 10 '⁶) is coated with a 0.5 μπι thick layer of TiN using conventional CVD technology at 930 ° C followed by a 2 μπι TiC_xNew layer using the MTCVD technique using T1Cl4, H2, N2 and CH3CN as process gases at a temperature of 885 ° C. In subsequent process steps during the same coating cycle, a layer of TiC precipitated_xO_z about 0.5 μπι thick at 1000 ° C using T1Cl4, CO and H2, and then the reactor was flowed with a mixture of 2% CO₂, 5% HCL and 93% H₂ for 2 min before a 2.2 μπι thick layer of 01-AI2O3 was precipitated. Process conditions during the deposition steps were as follows:

<td>Step</td><td>TiN</td><td>TLC_XNew</td><td>TiC_xO_z</td><td>flooding</td><td>A1₂O₃</td>

<td>TiCl₄</td><td> 1,5%</td><td> 1,4%</td><td> 2%</td><td></td><td></td>

<td>n₂</td><td> 38 %</td><td> 38 %</td><td></td><td></td><td></td>

<td>CO₂:</td><td></td><td></td><td></td><td> 2 %</td><td> 4%</td>

<td>CO</td><td></td><td></td><td> 6 %</td><td></td><td></td>

<td>AlCl3:</td><td></td><td></td><td></td><td></td><td> 3,2%</td>

<td>hrs₂s</td><td> -</td><td></td><td></td><td></td><td> 0,3 %</td>

<td>HC1</td><td></td><td> 1 %</td><td></td><td> 5 %</td><td> 3,2%</td>

<td>hrs₂^:</td><td>residual</td><td>residual</td><td>residual</td><td>residual</td><td>residual</td>

<td>CH3CN</td><td> -</td><td> 0,6 %</td><td></td><td></td><td></td>

12087swe_ny description.doc

528 696

<td>Print:</td><td>160 mbar</td><td>60 mbar</td><td>60 mbar</td><td>60 mbar</td><td>70 mbar</td>

<td>Temperature:</td><td>930 ° C</td><td>885 C</td><td>1000 ° C</td><td>1000 ° C</td><td>1000 ° C</td>

<td>Duration:</td><td>30 min</td><td>1 h</td><td>20 min</td><td>2 min</td><td>2 h</td>

X-ray diffraction analysis of the deposited AI2O3 layer showed that it consisted only of the α phase.

Example 2

Example 1 was repeated but the process time of TiC_xThe new step was 1.5 hours and the process time for the Al2C&gt; 3 step was 4 hours.

A coating of about 3 µm TiC_xNew and 4.5 µm (X-Al 2 O 3) were obtained.

Example 3

Example 1 was repeated, but the process time for the AI ​​2 O 3 step was 6 hours. A coating of about 2 µm TiC_xNew and 6.5 µm (X-Al 2 O 3) were obtained.

Example 4

Coated inserts from Examples 1, 2 and 3 were post-treated with wet blasting (both sides) and edge brushing under different conditions. Blasting was performed in two steps. First with a slurry containing Al2C&gt; 3 grains (grain size F80, FEPA standard) and water and then in a second step with Al2C&gt; 3 grains (grain size F320, FEPA standard). The purpose of these steps was to reduce the tensile stress level and to give the coating great smoothness. Four different blasting pressures 1.8, 2.0, 2.2 and 2.4 bar and two different spraying angles 45 ° and 90 ° were used in step 1. In step 2, only 2 bar pressure and 90 ° spray angle were used. Some inserts were also brushed with a nylon brush containing SiC grains to make a smooth coating surface along and near the cutting edge. The coatings for the various post-treated inserts were examined in a scanning electron microscope (SEM) at high magnification. It was clear from the investigation that only some inserts from Example 1 blasted at 2.4 bar showed sporadic flaking of the coating at the cutting edges. The inserts blasted at 90 ° angle showed a slightly better surface smoothness than those blasted at 45 ° angle.

The smoothness of the coating surface expressed as a well-known surface fineness value Ra was measured with AFM on a Surface Imaging System AG (SIS) equipment on all inserts except for the brushed and the non-blasted. Surface finesse was measured on ten randomly selected flat surfaces 12087swe_ny description.doc

528 696 (ΙΟμπιχΙΟμπι). The average value of these ten Ra values ​​was used as the surface fineness value, termed average Ra (MRA) in Table 1 below.

X-ray diffraction analysis using a Bragg-Brentano diffractometer, Siemens D5000, was used to determine the ratio 1 (012) / 1 (024) using CuKa radiation. The residual voltage was determined using ψ geometry on an X-ray diffractometer Bruker D8 Discover-GADDS equipped with laser video positioning, Euler l / 4 cradle, rotating anode as X-ray source (Cuk_Aradiation) and an area detector (Hi-star). A collimator of size 0.5 mm was used to focus the beam. The analysis was performed on TiC_xN_y(422) reflex using the goniometer substitutions 2θ = 126 °, ω = 63 ° and Φ = 0 °, 90 °, 180 °, 270 °. Eight ψ slopes between 0 ° and 70 ° were made for each Φ angle. Its²The i) / method was used to estimate the remaining voltage using the DIFFRAC software^plus Stress32 v. 1.04 from Bruker AXS with the constant E-module, E = 480 GPa and Poisson's ratio, v = 0.20 and the location of the reflex using the Pseudo-Voigt-Fit function. A biaxial stress state was confirmed and the mean value was used as the residual stress value.

Example 5

Cuts with different finishes were tested in two different milling operations, both with very demanding toughness properties. The following conditions were used:

Cutting test 1:

A milling operation in an alloy steel SS2541 was performed. The workpiece was a rectangular block. The cutter entered the workpiece a number of times from its long side, conditions classified as a difficult entrance.

Dry conditions Cutting speed Feed per tooth Axial cutting depth Radial cutting depth Number of teeth

V = 200 m / min Fz = 0.17 mm / Z Ap = 3 mm Ae = 16 mm = 1

12087swe_ny description.doc

528 696

Three inserts (one edge / insert) were run over the workpiece. The life span, expressed as the number of inputs that could be completed before breaking the edge is given in Table 1 below.

Cutting test 2:

A milling operation in an alloy steel SS2244 was performed. The shape of the workpiece was a thin long rod with a thickness much smaller than the diameter of the cutter. The cutter passed the rod longitudinally, conditions were classified as a difficult exit.

Wet conditions

V = 150 m / min Fz = 0.15 mm / Z Ap = 3 mm Ae = 7 mm = 2

Cutting speed Feed per tooth Axial cutting depth Radial cutting depth Number of teeth

3x2 inserts (one edge / insert) were run over the workpiece. The life span was expressed as the number of rods that could be completed before the rupture is given in the table below.

<td>Variant</td><td>Pressure / Blasting angle step 1</td><td>Average- Ra value, MRA</td><td>Sample 1 average food length</td><td>Sample 2 average food length</td><td>1 (012) / 1 (024) Al2O3 layer</td><td>Voltage in TiC_xN_y- layer</td>

<td>A</td><td>Non</td><td> &gt;&gt;0,09</td><td> 2</td><td> &lt; 1</td><td> 0, 8</td><td>700 MPa</td>

<td>B</td><td>brushed</td><td> —</td><td> 6</td><td> 5</td><td> 0,9</td><td>610 MPa</td>

<td>C</td><td>1.8 bar / 45 °</td><td> 0,11</td><td> 11</td><td> 5</td><td> 1,2</td><td>450 MPa</td>

<td>D</td><td>2.0 bar / 45 °</td><td> 0,10</td><td> 14</td><td> 11</td><td> 1,3</td><td>410 MPa</td>

<td>E</td><td>1.8 bar / 90 °</td><td> 0,09</td><td> 24</td><td> 17</td><td> 1,5</td><td>265 MPa</td>

<td>F</td><td>2.0 bar / 90 °</td><td> 0,09</td><td> 24</td><td> 19</td><td> 1,8</td><td>160 MPa</td>

<td>G</td><td>2.2 bar / 90 °</td><td> 0,06</td><td> 27</td><td> 22</td><td> 1,9</td><td>50 MPa</td>

<td>hrs</td><td>2.4 bar / 90 °</td><td> 0,07</td><td> 27</td><td> 23</td><td> 2,2</td><td>10 MPA</td>

<td>IN</td><td>2.2 bar / 90 °</td><td> 0,09</td><td> 18</td><td> 13</td><td> 1,8</td><td>280 MPa</td>

<td>J</td><td>2.4 bar / 90 °</td><td> 0,08</td><td> 10</td><td> 7</td><td> 2,0</td><td>390 MPa</td>

12087swe_ny description.doc

528 690

AH 2.5 μιη (TiN + TiC_xNew) +0.5 μιη TiC_xO_y + 2.2 μιη α-Α1₂03

I) 3.5 μιη (TiN + TiC_xNew) +0.5 μπι TiC_xO_y + 4.5 μιη α-Αΐ₂03

J) 2.5 μπι (TiN + TiC_xNew) +0.5 μιη TiC_xO_y + 6.6 μπι a-Al₂C&gt; 3

The results from cutting tests clearly show that the best toughness performance is achieved with the variants E, F, G and H which have the lowest tensile stresses in TiC_xThe new layer, the highest 1 (012) / 1 (024) ratios for Al₂O3 layer and low average Ra values. Variant I with a total coating thickness of 8.5 μη does not show as good performance (but still satisfactory) as the corresponding thinner, 5.2 μπι, variant G. Also variant J with a thick Al₂C&gt; 3 layers show a lower performance. Here are the tensions in TiC_xThe new layer is higher 390 MPa. These facts show that there is a certain parameter range of properties, which is directly linked to the life of the cutter. Accordingly, a number of conditions and properties must be met at the same time to achieve the high performance of the insert.

l2087swe_ny description.doc

528 696