Surface and Coatings Technology, 38 (1989) 339
MICROSTRUCTURE AND COMPOSITION OF PLASMA-NITRIDED Ti—6Al-4V LAYERS A. RAVEH Chemistry Division, NCR-Negev, P.O. Box 9001, Beer-Sheva (Israel) P. L. HANSEN Laboratory of Applied Physics I, Technical University of Denmark, 2800-Lyngby (Denmark) R.AVNI Chemistry Division, NRC-Negev, P.O. Box 9001, Beer-Sheva (Israel) A. GRILL Materials Engineering Department, Ben-Gurion University, Beer-Sheva (Israel) (Received
Summary Nitriding of titanium alloys was performed in an r.f. plasma of nitrogen, hydrogen and argon at a substrate temperature of about 500 °C.The microstructure and the composition of the nitrided layers obtained were studied by X-ray diffraction, scanning electron microscopy, transmission electron microscopy (TEM) and Auger electron spectroscopy (AES). Three phases were identified: (a) a solid solution of nitrogen in titanium, c~-(Ti,N), (b) tetragonal e-Ti2N and (c) cubic 5-TiN (NaC1-type structure). Two distinct layers were formed on top of the plasma-nitrided Ti—6A1—4V alloy, followed by a solid solution of nitrogen in the titanium alloy. The outer layer was identified as 5-TiN while the inner layer was identified as a mixture of s-TiN and c-Ti2N. The c-Ti2N was found to be highly oriented (002) while the 5-TiN was obtained as a randomly oriented polycrystalline layer. The outer 5-TiN was found to have a fine structure, while the inner 6-TiN plus e-Ti2N comprised large and oriented grains. TEM studies showed that the crystallite size in the upper layer was tens to hundreds of ~ngstroms, while in the inner layer it was one tenth to about half a micrometre. Energy-dispersive analysis performed with the TEM system and AES shows a variation in the bulk composition along the nitrided layer. The microstructures and the relative phase content are presented and discussed in relation to the gas feed composition in the plasma.
© Elsevier Sequoia/Printed in The Netherlands
1. Introduction The structure of the TiN phase is stable over a broad composition range and has a lattice parameter of 4.240 A at the stoichiometric composition . Deviations from the stoichiometric lattice parameters are usually related to stresses and impurities in the films [2, 31. Generally, TiN coatings are characterized by a preferred (111) orientation [21. The growth conditions affect orientation [41, microstructural features and film composition, i.e. the [N]/[Ti] ratio . Titanium or Ti—N phases can also have expanded lattice parameters caused by incorporation of interstitial nitrogen, leading to changes in the growth orientation. The microstructure of TiN films prepared by reactive r.f. sputtering [5, 6] shows a variation in grain size from 0.15 to 0.05 pm with an increase in the [N]/[Ti] ratio from 0.8 to 1.1. It is also reported  that a very finegrained structure with a grain size of about 120 A can be obtained by biasing the substrate. Grain sizes for stoichiometric films grown at 300 600 °Cby r.f. sputtering are normally about 1000 A, a value that decreases both if overstoichiometric films are grown and if high biases are used . The present work reports X-ray diffraction (XRD) and transmission electron microscopy (TEM) studies of the microstructure of the titanium alloy Ti—6Al—4V nitrided in r.f. plasmas of nitrogen, hydrogen and argon. -
2. Experimental details A detailed description of the r.f. plasma nitriding set-up and the determination of the composition of the samples (shown in Table 1) are given elsewhere . The grounded samples were placed in the plasma reactor at the centre of the r.f. coil. During the nitriding process the substrate temperature was 470 ±20 °C. TABLE 1 Chemical analysis of Ti—6Al--4V Elements
Al V Cu Fe Cr 0 C N H Ti
6.0 4.0 0.4max 0.3 max 0.1 max 0.2max 0.08 max 0.05 max 0.015 max Balance
10.0 3.6 0.3 0.2 0.1 0.6 0.03 0.2
Prior to the nitriding process, the samples were treated for 60 mm in a plasma with an Ar:H2 volume ratio of 2:1 at 250 ±10 C for sputter cleaning. The nitriding process was then performed in plasmas of different feed gas compositions; 8Ovol.%N2—2Ovol .%Ar, 80vol.%N2—lOvoI.%H2— 10 vol.%Ar, and nitrogen and hydrogen. The effect of the hydrogen concentration in the N2—H2 mixture on the microstructure of the nitrided layer was also studied. The microstructures of the nitrided films were examined with XRD with Cu Kct radiation and with a Philips 430 TEM operated at 300 kV acceleration voltage, which gives a point resolution of 2.3 A and a line resolution of 1.4 A suitable for high resolution studies. Two types of TEM specimens were prepared, i.e. cross-section and plane-view specimens. The cross-section specimen was cut perpendicular to the surface and ground to a thickness of about 100 pm. Two pieces were glued together on a copper ring (3 mm in diameter) and thinned further by argon sputtering. The plane-view specimen was cut parallel to the surface. The grinding was from the back (only the substrate) followed by “dimpling” and argon sputtering. In addition, the samples were examined by scanning electron microscopy (SEM), energy-dispersive analysis (EDS) performed in the TEM system, and by Auger electron spectroscopy and scanning Auger microscopy (AES—SAM) for composition measurements.
3. Results and discussions 3.1. X-ray diffraction In Fig. 1 the X-ray diffractogram of a sample nitrided in a 8Ovol.%N2— 2Ovol.%H2 plasma is compared with that of untreated Ti—6A1—4V. Additional peaks (marked with arrows) are observed in the nitrided sample compared with the untreated sample. An analysis of the peaks shows that the nitrided layer consists of three phases, namely (a) a solid solution of nitrogen in titanium, a-(Ti, N), (b) tetragonal e-T12N, and (c) cubic 6-TiN of NaC1-type structure. The study of the X-ray diffractograms as a function of the composition of the gas feed (N2—H2, N2—H2—Ar and N2—Ar) to the plasma shows a variation in the peak intensities of the e and 6 phases. This indicates a variation in the content of the phases and their orientation in the nitrided layer. Figure 2 shows XRD peak intensities of 6-TiN as a function of hydrogen concentration in the N2—H2 plasma. The relative XRD intensities in this figure are similar to the ASTM intensities of the 6 phase , indicating a randomly oriented phase. The variation in 1(200)/1(111) ratios of the 6 phase with the hydrogen concentration in the N2—H2 plasma implies a variation in the [N]/[Ti] ratios in the films . Figure 3(a) shows the XRD peak intensities of the c phase as a function of the hydrogen concentration in the N2—H2 plasma. The (002) peak is the
I. S-TIN ~5 E-TI 2N
diffraction angLe (20)
Fig. 1. X-ray diffractograms of untreated Ti—6A1--4V and of a sample nitrided in 8Ovol.%N2—20vol.%H2 plasma. Nitriding time, 5 h.
10 20 30 40 50 60 H2 Conc. (vol. %) Fig. 2. XRD peak intensities of the 6-TiN phase as a function of hydrogen concentration in the N2—H2 plasma.
strongest observed, whereas according to ASTM , (111) is the strongest peak for a random powder mixture. The intensities of the peaks in Fig. 3(a) indicate that addition of hydrogen to the nitrogen plasma causes the formation of the e phase with strong orientation in the (002) direction. The (002) orientation of the eTi2N phase was also obtained by conventional nitriding . However, in N2—Ar plasma the e phase obtained is oriented in the (111) direction. Only in the mixture N2—H2--Ar is the e phase obtained randomly oriented. This can be seen by comparing the peak intensities obtained in this case with the relative ASTM intensities , as shown in Fig. 3(b).
=~t 0 (a)
H2 Conc. (vol.%)
N2+H2 N2÷H2-i-Ar N2+Ar ASTM (b) Gas Composition Fig. 3. XRD peak intensities of the e-Ti2N phase as a function of (a) hydrogen concentration in the N2—H2 plasma, and (b) gas feed composition.
Figure 4 presents the phase content derived from the XRD peak patterns as correlated to the gas feed composition for a fixed concentration of 80 vol.% N2. This figure indicates that in samples treated in N2—Ar, equal amounts of 6, e and a phases are formed, while the concentration of the 6 phase increases to a maximum in samples treated in N2—H2 plasma and the e phase reaches its maximum in samples treated in N2—H2—Ar plasma. The phase content in the nitride layers essentially depends on the concentration of nitrogen in the films [14, 15], and the results in Fig. 4 indicate that at the highest concentration the hardest 6 phase is obtained in N2—H2 plasma. SEM studies revealed two distinct layers in the fractured film, as shown in Fig. 5. The outer surface layer (A) of the nitrided film has a fine columnar structure, while the inner layer (B) is formed of large oriented grains. Each of these layers was examined and identified by TEM. 3.2. Transmission electron microscopy Selected area electron diffraction (SAED) of the outer layer is shown in Fig. 6. The pattern corresponds to f.c.c. rings of pure cubic 6 phase TiN with
344 00 E
(;as Composition Fig. 4. Phase content as a function of gas feed composition.
400 311 220
Fig. h. Scanning electron micrograph of fractured film. Fig. 6. SAED of the upper layer of the nitrided film.
lattice parameter 4.24 A. Microdiffraction by a convergent electron beam with a spot of about 100 A shows a pattern of a single crystal of the 6 phase oriented in the  direction. Figure 7 is a TEM bright field micrograph of the inner layer. The figure shows crystals of approximately 0.1 pm in diameter with a lamellar structure. A similar structure has been observed in sputtered TiN films [6, 81. The SAED of the inner layer shows that it consists of mixed patterns of
F’ig. 7. TEIVI bright field micrograph of the inner layer (e plus
Fig. 8. SAED of the inner layer of the nitrided film showing 6-TiN plus
6-TiN and e-Ti2N phases. It can be seen in Fig. 8 that the (111) ring of the e phase was obtained between the (111) and (200) rings of the 6 phase. The single c phase is difficult to identify by SAED since its ring pattern is close
to those of the a phase and 6 phase. The TEM dark field images of the upper layer as shown in Fig. 9(a) indicate that the 6 phase has a grain size of tens to hundreds of ~ngstroms, while the grain size of the inner layer is one-tenth to about half a micrometre, as shown in Fig. 9(b). Microdiffraction in deeper regions shows a mixture of the a-Ti and e phase plus traces of the 6 phase, as shown in Fig. 10. Figure 10 shows the spot pattern from the a phase in the  direction, with the (111) ring pattern from the e phase. The spot pattern is somewhat distorted because the beam is not exactly along the  axis. TEM bright field micrographs of the boundary (marked with arrows) between the nitrided layer and the bulk are shown in Fig. 11(a). The high resolution image of the boundary layer (Fig. 11(b)) indicates that the crystallite size is similar to that in the inner layer, i.e. one-tenth to about half a micrometre. 3.3. EDS and AES—SAM analysis Figure 12 shows EDS performed in the TEM system in the upper layer and in the boundary between the nitrided layer and the bulk, compared with the untreated region. The analysis was performed for a sample treated in an 8Ovol.%N2—2Ovol.%H2 plasma. These analyses indicate that in the upper layer, the aluminium concentration decreases to one-tenth of its bulk value, while an increase in the concentrations of iron and vanadium is observed in the boundary layer between the bulk and the nitrided layer. In and between the two layers, clusters of small closely spaced precipitates (50 100 A in size) were observed, as shown in Fig. 13. Microdiffraction of the precipitates shows f.c.c. arid b.c.c. patterns corresponding to a-Fe and y-Fe. Furthermore, EDS showed that the precipitates contain aluminium, vanadium and chromium in addition to iron. -
r(I II) ct(0 13) 8(111)
Fig. 10. SAED of the deeper region of the nitrided film.
E.-fl2N +~ ON
a aB -
Fig. 11. Transmission electron micrographs of the boundary layer between the nitrided layer and the bulk: (a) bright field image; (b) high resolution image.
Figure 14 shows a typical SAM elemental profiling of a film nitrided in a plasma of 8Qvol.%N2—2Ovol.%H2 for 10 h. The SAM analysis was performed after polishing the sample at a tilt of 5~.Table 2 shows the analysis of the elements on the surface and at a depth of 20 pm. It can be seen that both the aluminium and vanadium concentrations decrease in the nitrided layer and increase at the boundary between the nitrided layer and the bulk. It can also be seen that the vanadium content increases as we penetrate deeper from the boundary into the bulk.
4. Conclusions (a) The nitrided Ti—6A1—4V film consists of two layers (6-TiN and 6TiN plus e-Ti2N) followed by an ct-(Ti, N) region. (b) A randomly oriented 6-TiN phase with fine microstructure was formed on top of the plasma-nitrided Ti—6Al—4V alloy. (c) The e-Ti2N was found to be highly oriented, depending on the gas mixture. (d) The crystallite size in the upper layer was hundreds of ângstroms, while in the inner layer it was about one-tenth to half a micrometre. (e) The aluminium and vanadium concentrations decreased in the upper layer, while the vanadium and iron concentrations increased in the boundary between the nitride layer and the bulk.
2. 00 8 7 4 C N
4. 00 1
6. 00 4 8 ~< E V
8. O~ 1O&//oh A
5 0 A X
Fig. 12. Energy-dispersive analysis: (a) untreated region; (b) in the upper layer; (c) in the boundary between the nitrided layer and the bulk.
(f) In and between the nitride layer ( and 6) and the substrate a-(Ti, N), clusters of small closely spaced iron precipitates were observed. Acknowledgment Professor A. Grill is the Incumbent of the Eric Samson Chair in Steel Processing.
50 am Fig. 1 3. Transmission electron micrograph ot the precipitates.
.~ . .—. i1~S
ic/S SIC 50142
Eig. 14. Peak-to-peak SAM elemental depth profiling of the nitrided film.
TABLE 2 Elemental analysis performed by SAM along the low angle (5°from surface) Element
Ti N Al
V Cr Fe
In the top layer
At 20 pm depth
47.5 47.2 3.0 1.0 0.8 0.5
70.2 ±1 14.8±1 10.0 ±1 5.0±1
References 1 L. E. Toth, Transition Metal Carbides and Nitrides, Academic Press, New York, 1971. 2 J. E. Sundgren, B. 0. Johnsson and S. E. Karlsson, Thin Solid Films, 105 (1983) 353. 3 J. Musil, L. Barados, A. Rajsky, J. Vyskocil, B. Dolegal, G. Loncar, K. Dadourex and V. Kubicek, Thin Solid Films, 136 (1986) 229. 4 H. M. Gabriel, in T. Takagi (ed.), Proc. mt. Ion Engineering Congr., Kyoto, September 12 16, 1983, Institute of Electrical Engineering of Japan, Tokyo, 1983, p. 1311. 5 J. E. Sundgren, B. 0. Johnsson, S. E. Karlsson and H. T. G. Hentzell, Thin Solid Films, 105 (1983) 367. 6 M. K. Hibbs, J. E. Sundgren, B. E. Jacobsson and B. 0. Johnsson, Thin Solid Films, 107 (1983) 149. 7 K. Y. Ahn, M. Wittmer and C. Y. Ting, Thin Solid Films, 107 (1983) 45. 8 J. E. Sundgren, Thin Solid Films, 128 (1985) 21. 9 A. Raveh, R. Avni and A. Grill, Thin Solid Films, (1989) to be published. 10 Powder Diffraction File for Inorganic Phases, Joint Committee for Powder Diffraction Standards, Swarthmore, PA, 1981, Card 6-0642. 11 G. Kimmel, Program XRIC (X-Ray Intensities Computations), Rep. 878, Philips Laboratory, Briarcliff Manor, NY, 1986. 12 Powder Diffraction File for Inorganic Phases, Joint Committee for Powder Diffraction Standards, Swarthmore, PA, 1981, Card 17-386. 13 K. T. Rie and Th. Lampe, Metalloberflhche, 37(1983) 1003. 14 C. J. Smithells (ed.), Metals Reference Book, Vol. II, Butterworth, London, 1967. 15 A. Raveh, G. Kimmel, U. Carmi, A. Inspektor, A. Grill and R. Avni, Surf. Coat. Technol., 36(1988) 183.