Coated cutting tool with h-AlN and Ti1-xAlxCyNz layers

Abstract

A coated cutting tool includes a substrate of cemented carbide, cermet, ceramics, steel or cubic boron nitride and a multi-layered wear resistant coating. The multi-layered wear resistant coating has a total thickness from 5 to 25 μm and includes refractory coating layers deposited by chemical vapour deposition (CVD) or moderate temperature chemical vapour deposition (MT-CVD). The multi-layered wear resistant coating has at least one pair of layers (a) and (b), with layer (b) being deposited immediately on top of layer (a). Layer (a) is a layer of aluminium nitride having hexagonal crystal structure (h-AlN) and a thickness from 10 nm to 750 nm. Layer (b) is a layer of titanium aluminium nitride or titanium aluminium carbonitride represented by the general formula Ti.sub.1-xAl.sub.xC.sub.yN.sub.z with 0.4≤x≤0.95, 0≤y≤0.10 and 0.85≤z≤1.15, having a thickness from 0.5 μm to 15 μm, and at least 90% of the Ti.sub.1-xAl.sub.xC.sub.yN.sub.z of layer (b) has a face-centered cubic (fcc) crystal structure.

Claims

1. A coated cutting tool comprising: a substrate of cemented carbide, cermet, ceramics, steel or cubic boron nitride; and a multi-layered wear resistant coating, wherein the multi-layered wear resistant coating has a total thickness from 5 to 25 μm and includes refractory coating layers deposited by chemical vapour deposition (CVD) or moderate temperature chemical vapour deposition (MT-CVD), and the multi-layered wear resistant coating includes at least one pair of layers (a) and (b) with layer (b) being deposited immediately on top of layer (a), wherein layer (a) is a layer of aluminium nitride having hexagonal crystal structure (h-AlN) and a thickness from 10 nm to 750 nm with a chlorine content of less than 5 at.-% and layer (b) is a layer of titanium aluminium nitride represented by the general formula Ti.sub.1-xAl.sub.xN.sub.z with 0.4≤x≤0.95, and 0.85≤z≤1.15, having a thickness from 0.5 μm to 15 μm, and at least 90% of the Ti.sub.1-xAl.sub.xN.sub.z of layer (b) has a face-centered cubic (fcc) crystal structure, wherein the multi-layered wear resistant coating includes a layer selected from the group of titanium nitride (TiN), titanium carbonitride (TiCN), and titanium aluminium nitride (TiAlN), said layer being deposited immediately underneath the first h-AlN layer of type (a) and having a columnar grain morphology.

2. The coated cutting tool of claim 1, wherein the multi-layered wear resistant coating includes one pair of layers (a) and (b).

3. The coated cutting tool of claim 1, wherein the thickness of the h-AlN layer(s) of the type (a) is from 10 nm to 400 nm.

4. The coated cutting tool of claim 1, wherein the thickness of the fcc-Ti.sub.1-xAl.sub.xN.sub.z layer(s) of type (b) is from 1 μm to 8 μm.

5. The coated cutting tool of claim 1, wherein the thickness ratio of the thickness of the h-AlN layer(s) of the type 1 (a) to the thickness of the Ti.sub.1-xAl.sub.xN.sub.z layer(s) of type (b) within each pair of layers (a) and (b) with layer (b) being deposited immediately on top of layer (a) is within the range from 0.01 to 0.5.

6. The coated cutting tool of claim 1, wherein the T.sub.1-xAl.sub.xN.sub.z layer(s) of type (b) within a pair of layers (a) and (b) with layer (b) being deposited immediately on top of layer (a) has a columnar grain morphology and a preferred crystallographic growth orientation characterized by a texture coefficient TC (111)>1.8, the TC (111) being defined as follows: $\mathrm{TC}\left(111\right)={\frac{I\left(111\right)}{I_0\left(111\right)}\left[\frac{1}{n}\underset{i=1}{\overset{n}{.Math.}}\frac{{I\left(\mathrm{hkl}\right)}_i}{{I_0\left(\mathrm{hkl}\right)}_i}\right]}^{-1}$ wherein I(111)=measured intensity of the (111) reflection I.sub.0(111)=standard intensity of the (111) reflection of the standard powder diffraction data according to the applied JCPDF-card no. 00-046-1200 I(hkl).sub.i=measured intensity of the (hkl).sub.i reflection I.sub.0(hkl).sub.i=standard intensity of the (hkl).sub.i reflection of the standard powder diffraction data according to the applied JCPDF-card no. 00-046-1200 n=number of reflections used in the calculation (here: n=4) (hkl).sub.i the (hkl).sub.i reflections used are: (111), (200), (220) and (311).

7. The coated cutting tool of claim 1, wherein the layer deposited immediately underneath the first h-AlN layer of the type (a) has a preferred crystallographic growth orientation characterized by a texture coefficient TC (200)>1.8, the TC (200) being defined as follows: $\mathrm{TC}\left(200\right)={\frac{I\left(200\right)}{I_0\left(200\right)}\left[\frac{1}{n}\underset{i=1}{\overset{n}{.Math.}}\frac{{I\left(\mathrm{hkl}\right)}_i}{{I_0\left(\mathrm{hkl}\right)}_i}\right]}^{-1}$ wherein I(200)=measured intensity of the (200) reflection I.sub.0(200)=standard intensity of the (200) reflection of the standard powder diffraction data according to the applied JCPDF-card no. 00-046-1200 when the said layer is a TiAlN or TiAlCN layer, and according to JCPDF card no. 00-42-1489 when the said layer is a TiN or TiCN layer I(hkl).sub.i=measured intensity of the (hkl).sub.i reflection I.sub.0(hkl).sub.i=standard intensity of the (hkl).sub.i reflection of the standard powder diffraction data according to the applied JCPDF-card no. 00-046-1200 when the said layer is a TiAlN or TiAlCN layer, and according to JCPDF card no. 00-42-1489 when the said layer is a TiN or TiCN layer n=number of reflections used in the calculation (here: n=4) (hkl).sub.i the (hkl).sub.i reflections used are: (111), (200), (220) and (311).

8. The coated cutting tool of claim 1, wherein the layer deposited immediately underneath the first h-AlN layer of the type (a) is a layer of titanium carbonitride (TiCN) and has a preferred crystallographic growth orientation characterized by a texture coefficient TC (422)>1.8, the TC (422) being defined as follows: $\mathrm{TC}\left(422\right)={\frac{I\left(422\right)}{I_0\left(422\right)}\left[\frac{1}{n}\underset{i=1}{\overset{n}{.Math.}}\frac{{I\left(\mathrm{hkl}\right)}_i}{{I_0\left(\mathrm{hkl}\right)}_i}\right]}^{-1}$ wherein I(422)=measured intensity of the (422) reflection I.sub.0(422)=standard intensity of the (422) reflection of the standard powder diffraction data according to the applied JCPDF-card no. 00-42-1489 I(hkl).sub.i=measured intensity of the (hkl).sub.i reflection I.sub.0(hkl).sub.i=standard intensity of the (hkl).sub.i reflection of the standard powder diffraction data according to the applied JCPDF-card no. 00-42-1489 n=number of reflections used in the calculation (here: n=5) (hkl).sub.i the (hkl).sub.i reflections used are: (111), (200), (220), (311) and (422).

9. The coated cutting tool of claim 1, wherein the layer deposited immediately underneath the first h-AlN layer of the type (a) is a layer of titanium aluminium nitride (TiAlN) or titanium aluminium carbonitride (TiAlCN) and has a preferred crystallographic growth orientation characterized by a texture coefficient TC (111)>1.8, the TC (111) being defined as follows: $\mathrm{TC}\left(111\right)={\frac{I\left(111\right)}{I_0\left(111\right)}\left[\frac{1}{n}\underset{i=1}{\overset{n}{.Math.}}\frac{{I\left(\mathrm{hkl}\right)}_i}{{I_0\left(\mathrm{hkl}\right)}_i}\right]}^{-1}$ wherein I(111)=measured intensity of the (111) reflection I.sub.0(111)=standard intensity of the (111) reflection of the standard powder diffraction data according to the applied JCPDF-card no. 00-046-1200 I(hkl).sub.i=measured intensity of the (hkl).sub.i reflection I.sub.0(hkl).sub.i=standard intensity of the (hkl).sub.i reflection of the standard powder diffraction data according to the applied JCPDF-card no. 00-046-1200 n=number of reflections used in the calculation (here: n=4) (hkl).sub.i the (hkl).sub.i reflections used are: (111), (200), (220) and (311).

10. The coated cutting tool of claim 1, wherein at least 95% of the Ti.sub.1-xAl.sub.xN.sub.z of layer (b) has a face-centered cubic (fcc) crystal structure.

11. The coated cutting tool of claim 1, wherein the multi-layered wear resistant coating includes a sequence of two or more pairs of layers (a) and (b).

12. The coated cutting tool of claim 11, wherein the sequence is 2, 3, 4, 5, 6, 7, 8, 9, or 10 pairs of layers (a) and (b).

Description

BRIEF DESCRIPTION OF THE FIGURES

(1) The figures show SEM microphotographs of the samples prepared according to the examples described below.

(2) FIGS. 1a, 1b show SEM microphotographs of a cross section (FIG. 1a) and of a top view onto the surface (FIG. 1b) of sample #1 (according to the invention);

(3) FIGS. 2a, 2b show SEM microphotographs of a cross section (FIG. 2a) and of a top view onto the surface (FIG. 2b) of sample #2 (according to the invention);

(4) FIGS. 3a, 3b show SEM microphotographs of a cross section (FIG. 3a) and of a top view onto the surface (FIG. 3b) of sample #3 (according to the invention);

(5) FIGS. 4a, 4b show SEM microphotographs of a cross section (FIG. 4a) and of a top view onto the surface (FIG. 4b) of sample #4 (comparative example);

(6) FIGS. 5a, 5b show SEM microphotographs of a cross section (FIG. 5a) and of a top view onto the surface (FIG. 5b) of sample #5 (according to the invention);

(7) FIGS. 6a, 6b show SEM microphotographs of a cross section (FIG. 6a) and of a top view onto the surface (FIG. 6b) of sample #6 (comparative example);

DEFINITIONS AND METHODS

(8) Fiber Texture and Texture Coefficient TC

(9) The term “fiber texture” or “texture”, respectively, as used herein and as it is generally used in connection with thin films produced by vapor deposition, distinguishes the orientation of the grown grains from random orientation. Three types of textures are usually distinguished in thin films and coatings: (i) random texture, when grains have no preferred orientation; (ii) fiber texture, where the grains in the coating are oriented such that one set of geometrically equivalent crystallographic planes {hkl} is found to be preferentially oriented parallel to the substrate, while there is a rotational degree of freedom of the grains around the fiber axis which is perpendicular to this plane, and thus preferentially orientated perpendicular to the substrate; and (iii) epitaxial alignment (or in-plane texture) on single-crystal substrates, where an in-plane alignment fixes all three axes of the grain with respect to the substrate.

(10) The crystallographic plane of a crystal is defined by the Miller indices, h, k, l. A means to express preferred growth, i. e. that one set of geometrically equivalent crystallographic planes {hkl} is found to be preferentially oriented parallel to the substrate, is the texture coefficient TC (hkl) calculated using the Harris formula on the basis of a defined set of XRD reflections measured on the respective sample. The intensities of the XRD reflections are standardized using a JCPDF-card indicating the intensities of the XRD reflections of the same material, e. g. TiCN, but with random orientation, such as in a powder of the material. A texture coefficient TC (hkl)>1 of a layer of crystalline material is an indication that the grains of the crystalline material are oriented with their {hkl} crystallographic plane parallel to the substrate surface more frequently than in a random distribution, at least compared to the XRD reflections used in the Harris formula to determine the texture coefficient TC.

(11) X-Ray Diffraction (XRD) Measurements

(12) X-ray diffraction measurements were done on a XRD3003 PTS diffractometer of GE Sensing and Inspection Technologies using CuKα-radiation. The X-ray tube was run in point focus at 40 kV and 40 mA. A parallel beam optic using a polycapillary collimating lens with a measuring aperture of fixed size was used on the primary side whereby the irradiated area of the sample was defined in such manner that a spill over of the X-ray beam over the coated face of the sample is avoided. On the secondary side a Soller slit with a divergence of 0.4° and a 25 μm thick Ni K.sub.β filter were used. Symmetrical θ-2θ scans within the angle range of 20°<2θ<155° with increments of 0.04° and 4 seconds counting time have been conducted. On the XRD raw data intensity corrections for thin film absorption were applied to all samples which take into account the limited thickness of the layer in contrast to the natural penetration depth in a bulk material. Furthermore an absorption correction was applied for samples where an additional layer was deposited above the layer for which the TCs were calculated. Finally Kα.sub.2 stripping (Rachinger method), background subtraction and a parabolic peakfit with 5 measuring points were applied. Therefore, any XRD peak intensities indicated herein mean the accordingly corrected intensities. For the calculation of the texture coefficients TC(hkl) a formalism proposed by Harris [Harris, G. B., Philosophical Magazine Series 7, 43/336, 1952, pp. 113-123] was applied. Herein the corrected net peak intensities I.sub.corr were correlated to the relative intensities I.sub.pdf taken from PDF-card 00-42-1489 for TiN or TiCN layers and from PDF-card 00-046-1200 for TiAlN or TiAlCN layers.

(13) $\mathrm{TC}\left(\mathrm{hkl}\right)={\frac{I\left(\mathrm{hkl}\right)}{I_0\left(\mathrm{hkl}\right)}\left[\frac{1}{n}\underset{i=1}{\overset{n}{.Math.}}\frac{{I\left(\mathrm{hkl}\right)}_i}{{I_0\left(\mathrm{hkl}\right)}_i}\right]}^{-1}$

(14) Transmission Electron Microscopy (TEM) EDS Analysis

(15) Transmission electron microscopic (TEM) analyses were performed in a FEI Titan 80-300 microscope with field emission cathode at an acceleration voltage of 300 kV. For EDS analyses an Oxford Inca EDS system was used. The preparation of samples for TEM was made by the in-situ lift-out technique using a combined FIB/SEM equipment to cut a thin cross sectional piece out of the surface and thin the sample down to sufficient electron transparency.

(16) Crystal Structure Determination by Electron Backscatter Diffraction (EBSD)

(17) The percentage of face-centered cubic (fcc) crystal structure of the Ti.sub.1-xAl.sub.xC.sub.yN.sub.z of layer (b) was determined by EBSD analysis on polished cross-sections of the samples. The polishing was done according to the following procedure: 6 min grinding using a grinding disc Struers Plano 220 and water; 3 min polishing using Struers 9 μm MD-Largo diamond suspension; 3:40 min polishing using Struers 3 μm MD-dac diamond suspension; 2 min polishing using Struers 1 μm MD-Nap diamond suspension; at least 12 min chemical polishing using Struers OP-S colloidal silica suspension with 0.04 μm average particle size. Prior to SEM/EBSD analysis the specimens were ultrasonically cleaned in ethanol and demagnetized. Inspection of the accordingly prepared specimens in the FE-SE (typically using an Everhart-Thornley secondary electron detector at 2.5 kV acceleration voltage and working distances of 3-10 mm) showed that grains of face centered cubic Ti.sub.1-xAl.sub.xC.sub.yN.sub.z layers are polished to a flat surface, showing a pronounced orientation contrast, whereas layers of h-AlN or h-AlN precipitated at grain boundaries of fcc-Ti.sub.1-xAl.sub.xC.sub.yN.sub.z layers are etched considerably stronger than fcc-phase grains, and therefore the surface of these proportions of the coating is lower than the fcc phase, and does not have a flat surface. Due to this topography, proportions in the coating which consist of h-AlN will give poor EBSD patterns in the EBSD analysis described below.

(18) EBSD analysis was performed in a Zeiss SUPRA40VP scanning electron microscope (SEM) with a field emission cathode using a 60 μm or 120 μm aperture and 15 kV acceleration voltage working in high current mode with a 70° incident angle of the electron beam to the polished samples surface at about 12 mm working distance. The EBSD detector was an EDAX Digiview camera, and the TSL OIM Data Collection 7 and TSL OIM Analysis 7 software packages were used for data collection and analysis, respectively. Typical acquisition and processing parameters for the EBSD maps are as follows: The map size was chosen with a length of about 25 μm parallel to the substrate surface and so that the entire thickness of the coating was covered using a ≤0.15 μm step size and a hexagonal grid of measurement points. A 4×4 or 8×8 binning and optionally a dynamic background subtraction is performed on the camera picture, using exposure times corresponding to 20 to 100 frames per second. However, as a rule, the preparation procedure described above yielded samples which gave diffraction patterns of the Ti.sub.1-xAl.sub.xC.sub.yN.sub.z layers with sufficient quality without performing background subtraction procedures. Indexing of the diffraction pattern is done by Hough transformation. The data points thus recorded should ideally be indexed with an average confidence index (CI) of >0.2. The CI is calculated by the TSL OIM Analysis 7 software during automated indexing of the diffraction pattern.

(19) In a first step, the EBSD map is cropped to get only the data points of the Ti.sub.1-xAl.sub.xC.sub.yN.sub.z layer (b) to be analysed. In a second step, grain CI standardization is carried out applying a grain tolerance angle of 5° and a minimum grain size of 5 data points. In a third step, partitioning of the so generated data set is carried out applying the filter CI>0.1, i.e. all data points that, after grain CI standardization, have a lower confidence index are disregarded. The ratio (number of data points indexed as fcc phase after CI standardization and filtering/total number of data points in the cropped map) corresponds to an area ratio of fcc phase within the Ti.sub.1-xAl.sub.xC.sub.yN.sub.z layer analysed (given in area-%). However, since pattern overlap and topography at grain boundaries lead to poor indexing of EBSD patterns obtained from fcc phase Ti.sub.1-xAl.sub.xC.sub.yN.sub.z, the values thus obtained represent a minimum fraction of fcc phase in the layer, the actual fraction being higher. Typically in Ti.sub.1-xAl.sub.xC.sub.yN.sub.z coatings where XRD and SEM gives no indication of h-AlN, and which therefore consist practically of about 100% fcc phase, the EBSD measurement and processing method described above yields >95 area-% of the EBSD map indexed as fcc phase.

(20) CVD Coatings

(21) The CVD coatings were prepared in a radial flow reactor, type Bernex BPX 325S, having 1250 mm height and 325 mm outer diameter. Gas flow over the charging trays was radial from a central gas tube.

EXAMPLES

(22) Sample Preparation

(23) For the preparation of cutting tools according to the present invention and of comparative examples cemented carbide cutting tool substrate bodies (composition: 90.5 wt-% WC, 1.5 wt-% TaC+NbC and 8.0 wt-% Co; geometry: SEHW1204AFN) were coated in a cylindrical CVD reactor, type Bernex BPX 325S, having a height of 1250 mm and a diameter of 325 mm.

(24) The gas flow over the substrate bodies was conducted radially from a central gas distribution tube, using a first and second precursor gas streams, PG1 and PG2. The first precursor gas stream, PG1, comprises, as far as required for the preparation of the desired coating, the gases AlCl.sub.3, TiCl.sub.4, CH.sub.3CN, N.sub.2 and hydrogen H.sub.2, and the second precursor gas stream, PG2, comprises, as far as required for the preparation of the desired coating, NH.sub.3 and H.sub.2. The first and second precursor gas streams, PG1 and PG2, were introduced into the reactor separately and combined immediately before entry into the reaction zone, i.e. after the outlet of the gas distribution tube.

(25) The experimental conditions for the deposition of the different layer types are given in the following table 1.

(26) TABLE-US-00001 TABLE 1 Deposition conditions for CVD layers Gas concentrations in precursor gas streams [Vol.-%] Temp. Pressure PG1 PG2 Layer type [° C.] [kPa] TiCl.sub.4 AlCl.sub.3 CH.sub.3CN N.sub.2 H.sub.2 NH.sub.3 H.sub.2 Ti.sub.1−xAl.sub.xN (1) 700 1 0.014 0.13 0 0 52 0.30 47.6 Ti.sub.1−xAl.sub.xN (2) 725 1 0.019 0.17 0 0 51.9 0.41 47.5 h-AlN (1) 700 0.3 0 0.24 0 0 47 1.14 51.6 h-AlN (2) 725 1 0 0.17 0 0 51.9 0.41 47.5 TiN (1) 790 2 0.44 0 0 43.3 32.5 2.16 21.6 TiN (2) 875 15 1.03 0 0 44 33 0 22 TiN (3) 850 15 1.03 0 0 44 33 0 22 TiCN (1) 825 7.5 1.77 0 0.47 10.9 65.2 0 21.7

(27) Coatings with different layer sequences were prepared. The following table 2 shows the layer types, layer sequences, deposition times, thicknesses, fiber textures and texture coefficients of the individual layers of samples according to the invention and of comparative samples.

(28) In table 2:

(29) L1=first layer immediately on substrate body surface

(30) L2=layer immediately underneath the first h-AlN layer (a)

(31) L(a)=Layer (a) of the type h-AlN

(32) L(b)=Layer (b) of the type Ti.sub.1-xAl.sub.xC.sub.yN.sub.z

(33) n.d.=not determined

(34) *=integral TC(hkl) for all Ti.sub.1-xAl.sub.xC.sub.yN.sub.z layers

(35) TABLE-US-00002 TABLE 2 Sample coatings according to the invention and comparative examples pairs of Sample # L1 L2 L(a) L(b) L(a) + L(b) 1 Layer type TiN (1) Ti.sub.1−xAl.sub.xN (1) h-AlN (1) Ti.sub.1−xAl.sub.xN(1) 1 (Inv.) Deposition time [min] 75 10 4 80 Layer thickness [μm] 3.5 0.7 0.2 5 Texture (hkl)/ {100} {100} EBSD — {111} TC TC(100) = 1.97 TC n.d. TC(111) = 3.27 2 Layer type TiN (2) Ti.sub.1−xAl.sub.xN (2) h-AlN (2) Ti.sub.1−xAl.sub.xN (2) 1 (Inv.) Deposition time [min] 120 70 2 10 Layer thickness [μm] 0.5 6 0.25 0.7 Texture (hkl) n.d. {111} — {111} TC(111) = 1.89* TC(111) = 1.89* 3 Layer type TiN (3) Ti.sub.1−xAl.sub.xN (2) h-AlN (2) Ti.sub.1−xAl.sub.xN (2) 7 (Inv.) Deposition time [min] 120 10 2 min/30 s/5 s/1 s/5 s/30 s/1 s 7 × 10  Layer thickness [μm] 0.4 0.6 0.15/0.05/<0.05/<0.05/ 7 × 0.6 <0.05/0.05/<0.05 Texture (hkl) n.d. {111} — {111} TC(111) = 1.98* TC(111) = 1.98* 4 Layer type TiN (3) Ti.sub.1−xAl.sub.xN (2) — — — (Comp.) Deposition time [min] 120 90 Layer thickness [μm] 0.4 7.5 Texture (hkl) n.d. {111} TC(111) = 2.97 5 Layer type TiN (2) TiCN (1) h-AlN (2) Ti.sub.1−xAl.sub.xN (2) 1 (Inv.) Deposition time [min] 120 45 10 45 Layer thickness [μm] 0.8 3 0.7 3.5 Texture (hkl) n.d. {422} — {111} TC(422) = 1.89 TC(111) = 3.05 6 Layer type TiN (2) TiCN (1) — Ti.sub.1−xAl.sub.xN (2) (Comp.) Deposition time [min] 120 45 45 Layer thickness [μm] 0.8 3 3.5 Texture (hkl) n.d. {211} {422} EBSD TC(422) = 3.72 TC n.d.

(36) EDS analysis were made on the TEM samples and for the AlN layer, Layer (a). No measurable content of W and Co was found in samples #1, #2 and #3. The Ti content was <0.3 at.-% in sample #1, 0.14 at.-% in #sample 2 and <0.5 at.-% in sample #3. The CI content was 2.9 at.-% in sample #1, 2.6 at.-% in #sample 2 and 2.2 at.-% in sample #3.

(37) Cuttinq Tests

(38) Milling operations using the cutting tool inserts of samples 1 to 6 were performed under the following conditions: Workpiece material: Steel DIN 42CrMo4 Coolant: None Feed per tooth: f.sub.z=0.2 mm Depth of cut: a.sub.p=3 mm Cutting speed: v.sub.c=283 m/min Setting angle: k=45°

(39) The development of the maximum flank wear, V.sub.Bmax, on the main cutting edge and the number of comb cracks were observed over a milling distance of 4000 mm in 800 mm steps. The following table 3 shows the development of V.sub.Bmax over the milling distance. In the milling test the cutting tool with the coating according to the present invention (sample 5) showed a significantly higher resistance against flank wear than the comparative example.

(40) TABLE-US-00003 TABLE 3 Cutting Test Results Milling Maximum Flank Wear V.sub.Bmax Distance Sample # 1 Sample # 2 Sample # 3 Sample # 4 Sample # 5 Sample # 6 [mm] (Inv.) (Inv.) (Inv.) (Comp.) (Inv.) (Comp.) 0 0 0 0 0 0 0 800 0.02 0.02 0.02 0.04 0.02 0.02 1600 0.04 0.02 0.03 0.04 0.04 0.04 2400 0.08 0.04 0.06 0.08 0.06 0.10 3200 0.12 0.08 0.08 0.10 0.10 0.18 4000 0.20 0.15 0.18 0.28 0.16 0.28