Surface hardening of Ti alloys by gas-phase nitridation: kinetic control of the nitrogen activity

FIELD OF THE INVENTION

The present invention relates to surface hardening of metals, and specifically to surface hardening of titanium and titanium alloys by the method of gas-phase nitridation with kinetic control of nitrogen activity.

BACKGROUND

Lightweight structural alloys based on titanium (Ti) have important aerospace and other industrial applications. For many such advanced applications, however, these alloys lack sufficient surface hardness, wear resistance, fatigue resistance, and corrosion resistance. A powerful approach to enhance their utility is to harden a thin layer below the surface of the material by inward diffusion of interstitial solutes, particularly carbon and nitrogen. This has long been recognized, and a number of treatments have been proposed over the years for such “case hardening” (surface hardening) of Ti-base alloys, including nitridation, carburization, and oxidation by gas-discharge plasmas, References 1-8 high-energy ion implantation, References 9,10 exposure to molten salts, References 11,12 electrochemical hydrogenation, References 13,14 or laser treatment (surface melting in nitrogen or carbon-containing atmospheres). References 1,15-18 However, those approaches invariably cause precipitation of carbides and nitrides, predominantly at the surface and at grain boundaries. This decreases the concentration of the desired interstitial solute, and the presence of titanium carbides and nitrides compromises the fatigue resistance. Moreover, a second phase at the alloy surface may reduce the corrosion resistance. Therefore, it is important that the interstitial solutes stay in solid solution and do not form second phases. Ideally, solute concentration profiles should decrease smoothly with increasing distance from the surface, in order to avoid discontinuities of mechanical or electrochemical properties that might cause spallation or corrosion.

It is true that even when nitride or carbides precipitates form near the surface, a homogeneous solid solution with the desired properties may still be established somewhat deeper below the surface. Unfortunately, removal of the layer containing the precipitates generally introduces substantial surface roughness or cracks, negatively impacting the mechanical and corrosion behavior in a similar fashion as the presence of carbide or nitride particles.

REFERENCES

1. T. Bell, H. W. Bergmann, J. Lanagan, P. H. Morton, and A. M. Staines, Surface Engineering 2, 133 (1986).

2. T. Sato and K. Akashi, Journal of Japan Institute of Light Metals 42, 650 (1992).

3. N. Yasumaru, K. Tsuchida, E. Saji, and T. Ibe, Journal of the Japan Institute of Metals 56, 104 (1992).

4. N. Yasumaru, K. Tsuchida, E. Saji, and T. Ibe, Materials Transactions, JIM 34, 696 (1993).

5. H. J. Spies, B. Larisch, K. Hoeck, E. Broszeit, and H. J. Schroeder, Surface and Coatings Technology 74-75, 178 (1995).

6. H. Dong, A. Bloyce, and T. Bell: in Surface Performance of Titanium, J. K. Gregory, J. J. Rack, and D. Eylon, cds., TMS, Warrendale, Pa., 1996. pp. 23-31.

7. T. A. Panaioti, Metallovedenie i Termicheskaya Obrabotka Metallov 40, 32 (1998).

8. Z. Okamoto, H. Hoshika, and M. Yakushiji, in Technology Reports of Kansai University (Kansai Univ, 2001), pp. 167-77.

9. A. Bloyce, Proceedings of the Institution of Mechanical Engineers, Part J (Journal of Engineering Tribology) 212, 467 (1998).

10. J. L. Viviente, A. Garcia, A. Loinaz, F. Alonso, and J. I. Onate, Vacuum 52, 141 (1999).

11. F. D. Lai, T. I. Wu, and J. K. Wu, Surface and Coatings Technology 58, 79 (1993).

12. T. I. Wu and J. K. Wu, Surface and Coatings Technology 90, 258 (1997).

13. T. I. Wu and J. K. Wu, Scripta Metallurgica et Materialia 25, 2335 (1991).

14. T.-I. Wu and J.-K. Wu, Metallurgical Transactions A (Physical Metallurgy and Materials Science) 24, 1181 (1993).

15. H. M. Flower, A. Walker, and D. R. F. West, Scripta Metallurgica 19, 923 (1985).

16. M. Okutomi, A. Obara, K. Tsukamoto, and H. L. Shen, Proceedings of the SPIE—The International Society for Optical Engineering 3550, 229 (1998).

17. M. Okutomi, A. Obara, and K. Tsukamoto: Proc. High Temperature Coatings III, San Diego, Calif., 1999, pp. 341-50.

18. Z. Song, Z. Chunhua, W. Weitao, and W. Maocai, Acta Metallurgica Sinica 37, 315 (2001).

19. J. Bars, D. David, E. Etchessahar, and J. Debuigne, Metallurgical Transactions A 14, 1537 (1983).

20. E. Metin and O. T. Inal, Metallurgical Transactions A (Physical Metallurgy and Materials Science) 20, 1819 (1989).

21. J. Crank, The mathematics of diffusion 61,62 (Clarendon Press, Oxford, 1956).

22. W. Boettinger, M. Williams, S. Coriell, U. Kattner, and B. Mueller, Metallurgical and Materials Transactions B: Process Metallurgy and Materials Processing Science 31, 1419 (2000).

23. A. Rosen and A. Rottem, Materials Science and Engineering 22, 23 (1976).

24. S. Fujishiro and D. Eylon, Scripta Metallurgica 11, 1011 (1977).

25. M. A. Daeubler and D. Helm, in Proceedings of the International Conference on Titanium Products and Applications (Orlando, Fla., USA, 1990), pp. 224-255.

26. D. Satyanarayana and M. Pandey, Scripta Metallurgica et Materialia 25, 2273 (1991).

27. L. F. S. Dumitrescu, M. Hillert, and B. Sundman, Zeitschrift für Metallkunde 90, 534 (1999).

28. S. Jonsson, Zeitschrift für Metallkunde 87, 691 (1996).

ASPECTS OF THE INVENTION

It is an aspect of the present invention to provide methods and apparatus as defined in one or more of the appended claims and, as such, having the capability of accomplishing one or more of the following subsidiary objects.

Another aspect of the present invention is to provide a method for the surface hardening of titanium and titanium alloys by means of nitridation while minimizing the potential for the formation of metal nitrides and the precipitation thereof, predominantly at surfaces and grain boundaries.

Yet another aspect of the present invention is to provide a gas-phase method for the surface hardening of titanium and titanium alloys by means of nitridation while minimizing the potential for the formation of metal nitrides and the precipitation thereof, predominantly at surfaces and grain boundaries.

Still another aspect of the present invention is to provide a gas-phase method for the surface hardening of titanium and titanium alloys by means of nitridation while minimizing the potential for the formation of metal nitrides and the precipitation thereof, predominantly at surfaces and grain boundaries such that said method can be applied to a titanium-based alloy within a period of hours or tens of hours.

A further aspect of the present invention is to provide a gas-phase method for the surface hardening of titanium and titanium alloys by means of nitridation, said method being adaptable to industrial scale-up.

SUMMARY OF THE INVENTION

The present invention is a method for surface hardening of metal surfaces by means of nitridation, said method being characterized by the exposure of the metal surface to very low activity nitrogen gas at elevated temperature. In particular, the invention is a method for nitridation of titanium and titanium alloy surfaces.

The present invention is directed to a method for surface hardening of a metal surface of a work piece formed of a metal from the group consisting essentially of titanium and titanium alloys. The method comprises the steps of:

heating the work piece to a desired temperature of less than about 1000° C.;

exposing the metal surface to a nitrogen gas having a partial pressure lower than about 10⁻²Pa; and

maintaining the work piece in the nitrogen gas for a period of time so that nitrogen is absorbed onto the surface of the work piece and diffused into the work piece for a desired distance while preventing the formation of nitrides on the surface of the work piece and within the work piece.

Also according to the present invention the method includes the step of exposing the metal surface to a nitrogen gas having a partial pressure of between about 10⁻⁷Pa and 10⁻²Pa and a temperature of less than about 1000° C. More particularly, the metal surface can be exposed to a nitrogen gas having a partial pressure of between about 10⁻⁵Pa and 10⁻³Pa and at a temperature of between about 700° C. and 1000° C. Still more particularly, the metal surface can be exposed to a nitrogen gas having a partial pressure of between about 10⁻⁵Pa and 10⁻³Pa and at a temperature of about between about 800° C. and 900° C.

According to the invention, the metal surface can be exposed to the nitrogen so that it is diffused into work piece for a desired distance of about 2 and 250 μm. This can be accomplished by maintaining the work piece in the nitrogen gas for a period of time of from about 1 to 250 hours.

The present invention is further directed to a method for surface hardening of a metal surface of a work piece formed of a metal from the group consisting essentially of titanium and titanium alloys. The method comprises the steps of:

heating the work piece to a first temperature of less than about 1000° C.;

exposing a surface of the work piece to a nitrogen gas whose pressure is determined by a second temperature to which a first powder of metal nitride and a second powder of the corresponding metal forming the metal nitride is exposed; and

Also according to the present invention the method includes the step of exposing the surface of the work piece to a nitrogen gas having a pressure of between about 10⁻⁷Pa and 10⁻²Pa. More particularly, the method includes the step of exposing the surface to a nitrogen gas having a pressure of between about 10⁻⁷Pa and 10⁻²Pa created by the decomposition of the nitride powder and simultaneously reacting the metal powder with the nitrogen to form metal nitride. Still more particularly, the method includes the step of exposing the metal surface to a nitrogen gas having a partial pressure of between about 10⁻⁵Pa and 10⁻³Pa and at a second temperature of between about 700° C. and 1000° C. The method also includes the step of exposing the metal surface to a nitrogen gas having a partial pressure of between about 10⁻⁵Pa and 10⁻³Pa and at a second temperature of about between about 800° C. and 900° C.

Further, according to the present invention the method includes the step of exposing the metal surface so that nitrogen is diffused into the surface of the work piece for a desired distance of about 2 and 250 μm. The method also includes the step of encapsulating the work piece together with powders of a metal nitride and the corresponding metal in an evacuated inert container, such as an ampoule of fused silica.

Further, according to the present invention the surface of the work piece is exposed to a nitrogen gas at one or more pressures different than the first pressure.

BRIEF DESCRIPTIONS OF THE FIGURES

The structure, operation, and advantages of the present preferred embodiment of the invention will become further apparent upon consideration of the following description taken in conjunction with the accompanying drawings, wherein:

FIG. 1 shows the Ti—N equilibrium phase diagram.

FIG. 2 shows hardness of titanium as a function of the nitrogen concentration in solid solution.

FIG. 3 shows the theoretical case depth at 860° C. as a function of time.

FIG. 4 is a schematic illustration of the concept of surface nitridation of titanium and titanium alloys by the method of kinetic control.

FIG. 5 illustrates, schematically, the powder-pack method for nitriding titanium and titanium alloys in a non-nitride-forming gas atmosphere.

FIG. 6 illustrates thermodynamic data indicating the equilibrium dissociation pressure of nitrogen for various metal nitrides over a wide range of temperatures.

FIG. 7 is a schematic model for determining the suitable range of nitrogen partial pressures for nitridation of titanium and titanium alloys.

FIG. 8 illustrates theoretical conditions for gas-phase nitridation of titanium alloys at 860° C.

FIG. 9 is a diagrammatic summary of calculated nitrogen concentration profiles for pure titanium exposed to a nitrogen partial pressure of 10⁻⁴Pa for four different time periods.

FIG. 10 is an illustration of Ti-6Al-4V specimens treated by power pack nitridation under different conditions.

FIG. 11 shows X-ray diffractograms for Ti-6Al-4V after nitridation with various powder packs.

FIG. 12 shows X-ray diffractograms of Ti-6Al-4V demonstrating the best results obtained by powder-pack treatment with Si₃N₄—Si.

FIG. 13 which includes FIGS. 13a and 13b shows XPS spectra for as-received Ti-6Al-4V and for a Ti-6Al-4V specimen after nitridation and after polishing 3 μm of material off the surface.

FIG. 14 contains scanning electron microscopy images of Vickers hardness indents in nitrided Ti-6Al-4V.

FIG. 15 is a bar graph display of hardness data related to Ti-6Al-4V after nidtridation in Si₃N₄powder packs, compared to as-received untreated material.

FIG. 16 is a schematic diagram of an improved processing scheme for “clean” powder-pack nitridation.

FIG. 17 shows X-ray diffractograms obtained from Ti-6Al-4V nitrided for 24 h at 860° C. in “clean” CrN—Cr power packs held at various temperatures.

FIG. 18 shows X-ray diffractograms of an annealed specimen of Ti-6Al-4V compared with a specimen nitrided with a “clean” CrN—Cr powder pack maintained at 676° for 24 hours, then at 650° C. for 24 hours, and finally at 625° C. for 24 hours.

FIG. 19 shows tensile stress-strain curves of an untreated, an annealed, and 2 nitrided samples of Ti-6Al-4V.

FIG. 20 shows two TEM images illustrating dislocation and twin boundaries in a nitrided sample.

DETAILED DESCRIPTION IN THE INVENTION

The present invention, relates to an a process for improving the surface properties of titanium alloys through inward diffusion of interstitial solutes without the formation of any second phases.

The Ti—N equilibrium phase diagram in FIG. 1 indicates considerable solubility of nitrogen in pure titanium. At 1050° C., the solubility reaches its maximum of 23 atomic % (at %) in titanium (Ti). Correspondingly high solubilities exist in Ti-based alloys. However, because unalloyed titanium undergoes an allotropic phase transformation from α-Ti to β-Ti at 882° C., which is usually undesired in engineering materials, the highest possible nitridation temperature for two-phase alloys, such as Ti-6Al-4V, is probably lower. At 860° C., for example, the nitrogen solubility in a-titanium is still about 14 at %.

According to published data (Reference 19), as shown in FIG. 2, a nitrogen content of 14 at % increases the hardness of α-titanium from about 375 HV to about 900 HV. Moreover, the diffusion coefficient of nitrogen in titanium at 860° C. is sufficiently high (1.22×10⁻¹⁰m²s⁻¹)²⁰to obtain a case depth d(=(2Dt)^1/2) of about 50 μm after reasonably short treatment times—on the order of 24 h, as shown in FIG. 3. In the past, therefore, experiments were carried out to nitride titanium alloys of various compositions with gas mixtures containing between 1 and 100% nitrogen at atmospheric pressure. However, titanium nitrides inevitably formed because the affinity of titanium for nitrogen is so high and titanium nitrides are so stable that their formation could not have been suppressed under the conditions of those experiments.

In fact, equilibrium thermodynamic analysis shows that the formation of titanium nitrides cannot be avoided unless the nitrogen activity is kept extremely low. For temperatures below 860° C., the nitrogen partial pressure in equilibrium needs to be below 10⁻¹⁵Pascals Pa [p *]. Accordingly, the nitrogen partial pressures that can be realized in a nitrogen-containing gas under atmospheric pressure are many orders of magnitude too high. Even nominally “pure” Ar, often used to generate an inert atmosphere, contains too high a nitrogen partial pressure to avoid nitride precipitation.

In reality, somewhat higher nitrogen partial pressures than p* can be used because the nitrogen atoms impinging on the specimen surface diffuse into the material, thus rendering the surface nitrogen concentration too low for nitride formation. Assuming that every nitrogen atom impinging on the surface will diffuse inwards, nitride formation is avoided as long as the diffusion current density in the solid,
$\begin{matrix} j_{s} = - {D_{S} (\frac{\partial c_{S} [x, t]}{\partial x})}_{x = 0}, & (1) \end{matrix}$

is larger than the flux density, j_g, of nitrogen atoms impinging on the surface. In eqn. (1), x is the spatial coordinate into the specimen (the surface is at x=0), t the time, c_sthe concentration of interstitial nitrogen atoms in the solid, and D_stheir diffusion coefficient. In order to build up a “case” with increased nitrogen concentration, j_gmust be larger than j_s. This concept is illustrated in FIG. 4. According to the kinetic theory of an ideal gas,
$\begin{matrix} j_{g} = \frac{1}{6} c_{g} v_{g} = \frac{1}{6} c_{g} \sqrt{\frac{3 k_{B} T}{m}} = \sqrt{\frac{c_{g}}{12 m} p,} & (2) \end{matrix}$

where c_s=c_s[x,t] and c_g=constant are the nitrogen concentrations in the solid and in the gas, respectively, v_gthe average velocity of the gas molecules, m the mass of the gas molecules, k_Bthe Boltzmann constant, T the absolute temperature, and p the nitrogen partial pressure. Therefore, the nitrogen partial pressure p required for case hardening needs to fulfill the condition
$\begin{matrix} p \overset{>}{\approx} \sqrt{12 m k_{B} T} \cdot {D_{S} (\frac{\partial c_{S}}{\partial x})}_{x = 0} . & (3) \end{matrix}$

Numerically solving the diffusion equation
$\begin{matrix} \frac{\partial c_{S}}{\partial t} = D_{S} \frac{\partial^{2} c_{S}}{\partial x^{2}} & (4) \end{matrix}$

for a concentration-independent diffusion coefficient (Fick's 2nd law), Reference 21 we find that the suitable range of nitrogen partial pressures for nitridation at a temperature of between about 800° C. and 900° C. and preferably about 850° C. and 870° C. and most preferably about 860° C. absent nitride formation is between about 10⁻⁵and 10⁻³Pa. Throughout the present specification, a temperature of 860° C. is often specified since the testing procedure was done on a titanium alloy workpiece where this temperature was suitable. However, it is within the terms of the invention that the recitation of a temperature of 860° C. is defined herein as “a temperature of less than about 1000° C. and more particularly between about 700° C. and 1000° C. and preferably about 800° C. and 900° C. and most preferably about 825° C. and 875° C.”. Further, the partial pressure for nitridation absent nitride formation is between about 10⁻⁷and 10⁻²Pa and preferably between about 10⁻⁵and 10⁻³Pa.

The implications of this result are striking. Because of the limited number of gas molecules impinging on the surface of a titanium alloy at temperatures and nitrogen activities below 10⁻¹⁵Pa, preferably in the range where Ti₂N or TiN could form from a thermodynamic viewpoint, the kinetics of nitrogen adsorption and inward diffusion can be rapid enough to prevent the formation of surface or interior nitrides, thereby obviating the usual problems associated with nitridation of Ti alloys. In spite of the simplicity of this argument, we know of no prior work that recognizes that the kinetics of nitrogen adsorption from a nitrogen-containing gas phase can be exploited to prevent nitride formation, while still allowing significant solid-solution hardening by interstitially dissolved nitrogen.

While the kinetic conditions just described significantly relax the requirement of 10⁻¹⁵Pa obtained by equilibrium thermodynamics, the nitrogen partial pressure still must be controlled at an extremely low level above 10⁻⁷Pa and in a preferred range of between about 10⁻²and 10⁻⁷more particular in a preferred range of between about 10⁻³and 10⁻⁵and still more particularly below 10⁻³Pa. To deal with this requirement, we have developed an innovative processing techniques, employing a powder pack (the so-called “pack cementation” process).

With the powder-pack technique, we can generate a well-defined nitrogen partial pressure by encapsulating the work piece or specimen in an evacuated ampoule of fused silica (or other inert containers), together with powders of a metal nitride and the corresponding metal as shown in FIG. 5. By heating the powder pack to an appropriate temperature T_p, which can differ from the temperature, T_sof the specimen, nitrogen is released by decomposition of the nitride powder. At the same time, the metal powder reacts with the nitrogen to form the metal nitride. In equilibrium, and as long as a sufficient powder is provided, the buffering action of the two powders provides a partial pressure of nitrogen in the chamber that corresponds to the dissociation pressure of the metal nitride at the temperature of the pack.

The formation of titanium nitrides can be avoided by operating with a TiN—Ti powder pack and setting T_pto equal or substantially equal T_s. In this case, the nitrogen partial pressure corresponds to the equilibrium dissociation pressure of TiN at the specimen, and thus to a nitrogen partial pressure p just below the pressure at which TiN begins to form. However, the dissociation pressure of TiN is extremely low (<10⁻¹⁵Pa). Therefore, operating with TiN—Ti equilibrium conditions will produce an extremely small flux density of nitrogen atoms impinging on the specimen surface. This will not be sufficient to obtain significant case hardening in realistic times, particularly since the inward diffusion of nitrogen will be rapid compared to the rate of arrival at the surface.

As mentioned before, the result of the analysis presented above suggests that the preferred range of nitrogen partial pressure p for T_s=860° C. (a favorable specimen temperature) is between 10⁻⁵and 10⁻³Pa. The thermodynamic data in FIG. 6, obtained from the JANAF tables, indicate the nitrogen partial pressures that can be obtained with a variety of metal-metal-nitride combinations over a broad interval of temperature. The combinations CrN—Cr, VN—V, and Si₃N₄—Si all have dissociation pressures in the desired range. As mentioned before, it is possible to fine-tune the nitrogen partial pressure by adjusting the temperature of the powder pack to be different from the temperature of the specimen. Once the material of the specimen is selected, the partial pressure of the nitrogen can be determined along with the temperature from the thermodynamic data of the type in FIG. 6,

In order to determine the suitable range of nitrogen partial pressures for nitridation at 860° C., we modeled diffusion of nitrogen into an infinite Ti plate of thickness 2 h. The specimen, in the form of a plate, was assumed to lie parallel to the y-z plane of the reference frame, with the upper and lower specimen surfaces intersecting the x-axis at x=−h and x=+h, respectively (see FIG. 7). A further assumption was that the nitrogen atoms impinge on the specimen surface with a homogeneous flux rate Γ, and that each atom impinging on the specimen surface will diffuse into the material. Solving the diffusion equation (4) under these boundary conditions with a concentration-independent diffusion coefficient (see Reference 20)
$\begin{matrix} D_{N} = D_{0} Exp [\frac{Q_{a}}{k_{B} T}] & (5) \\ D_{0} = (0.96 \pm 0.08) \times 10^{- 4} m^{2} s^{- 1} & (6) \\ Q_{a} = (2.22 \pm 0.02) e V & (7) \end{matrix}$

according to Fick's 2^ndlaw yields the solution (see Reference 21)
$\begin{matrix} C = C_{0} + \frac{Γ h}{D} (\begin{matrix} \frac{Dt}{h^{2}} \div \frac{3 x^{2} - h^{2}}{6 h^{2}} - \frac{2}{π^{2}} \times \\ \sum_{n = 1}^{\infty} \frac{{(1)}^{n}}{n^{2}} Exp [- \frac{{Dn}^{2} π^{2} t}{h^{2}}] Cos [\frac{n π x}{h}] \end{matrix}) . & (8) \end{matrix}$

Numerically solving eqn. (8) for a plate thickness of 2 h=2 mm (a quasi-infinite thickness under the conditions considered here), the nitrogen mole fraction X_Nwas obtained as a function of specimen depth (see FIG. 7) for a variety of temperatures and nitrogen partial pressures p. From the data base of concentration profiles obtained in this way, the three curves displayed in FIG. 8 were extracted, all or which refer to a temperature of 860° C. Curve (a) shows the nitrogen partial pressure required in the gas phase for obtaining a surface concentration of 14 at % nitrogen, the equilibrium solubility limit of pure Ti at 860° C., as a function of time. Conversely, this curve shows the processing time it takes to obtain a surface concentration of 14 at % nitrogen for a given nitrogen partial pressure. At a pressure of 2.5×10⁻⁴Pa, for example, it will take about 20 hours (h) to increase the surface concentration of nitrogen to the equilibrium solubility limit of 14 at %. According to FIG. 8, nitridation at 860° C. should work best with a nitrogen partial pressure between about 10⁻⁵and 10⁻³Pa. Curves (b) and (c) indicate the nitrogen mole fraction at a depth of 50 and 100 μm below the specimen surface, respectively, for the times and pressures given by curve (a). For example, if the nitrogen partial pressure is such that the nitrogen solubility limit at the alloy surface is reached after 100 h, the nitrogen concentration at a depth 50 μm below the surface will reach about 7.0 at %, while at 100 μm below the surface, it will reach about 2.5 at %.

The results for a particularly suitable set of parameters are shown in FIG. 9. Here, it was assumed that the titanium was treated with a nitrogen partial pressure of about 10⁻⁴Pa at 860° C. for times between about 10 and 300 h. The results demonstrate that treatments between 30 and 100 h will raise the surface concentration to about 20 at %, near the solubility limit indicated by FIG. 1, and that a case depth of about 50 to 100 μm should be achievable.

All our powder pack experiments have been carried out on Ti-6Al-4V (grade 5) alloys. The specimens we used in our experiments were discs about 10 mm in diameter and about 4 mm thick, cut from bar stock (FIG. 10). Prior to nitridation, the surfaces of the specimens were polished with 4000 grit silicon carbide paper.

Initially, the experiments were conducted with three different powder packs: VN—V, CrN—Cr, and Si₃N_4-Si. Initial work established appropriate techniques for cleaning the interior of fused silica ampoules, handling of powders, evacuating ampoules containing the samples and powders with a rotary pump without disturbing the powder pack, and sealing the ampoules by means of a hydrogen flame. In search of suitable nitridation conditions, a series of experiments were run in which the work piece and the powder mixture were encapsulated in 100 mm long ampoules. These ampoules were annealed in a tube furnace at temperatures up to 860° C. for time periods ranging between about 10 and 100 h.

In order to enable experiments with the flexibility of different temperature for the specimen and powder pack, the short ampoules were replaced with 1 meter long ampoules. In these ampoules the workpiece was placed at one end and the powder at the other. Each end was then placed in one of two adjacent tube furnaces, such that the specimen was kept at 860° C. and the power pack at a lower temperature, typically between about 350° C. and 800° C.

After the treatment, the specimens were inspected visually. Comparison with the results obtained by X-ray diffraction revealed that a golden surface indicated the presence of nitrides, while a black surface indicated the presence of oxides. The occurrence of either surface color immediately indicates that the processing conditions were not appropriate. FIG. 10 presents diagrammatically examples of specimens exhibiting nitride and oxide scales, along with a specimen whose surface remained metallic looking. That us, in FIG. 10, there is a diagram of Ti-6Al-4V specimens treated by power pack nitridation under different conditions wherein (a) Cr—CrN, 48 h at 860° C., is shiny; (b) Si—Si₃N₄, 48 h at 860° C., has a nitride scale; (c) Al₂O₃, 48 h at 860° C., has an oxide scale; and (d) Si—Si₃N₄, 48 h at 700° C., has a nitride scale.

All specimens were examined before and after nitridation by X-ray diffractometry (XRD) with monochromated Cu≠Kα radiation. Normal-incidence diffractograms were recorded in the standard Bragg-Brentano geometry, typically in the range 20°<2Θ<80°,with steps of 0.02° and counting times of 1.2 or 10 s/step. These diffractograms sample the first 5 μm below the specimen surface. We have also recorded diffractograms under glancing incidence. In these measurements, the primary beam makes a constant angle of only about 1° with the specimen surface. As a result, glancing-incidence diffractograms sample roughly only about the first 0.05 μm below the specimen surface.

FIG. 11 presents normal-incidence diffractograms of Ti-6Al-4V after nitridation for 48 h at 860° C. with various powder packs, including Ti, Cr, V, and Si powder packs.

The results of these experiments, as documented by FIG. 11, can be summarized as follows:

1. A Ti—TiN powder pack has hardly any effect on the workpiece. The diffractogram exclusively exhibits peaks of titanium (the peak at 39.5° 2θ originates from β-titanium) with no notable peak shifts that would indicate a significant lattice expansion by uptake of nitrogen. Accordingly, the dissociation pressure of TiN at 860° C. is too low.

2. The presence of a Cr/CrN powder pack at 860° C. completely transforms the titanium within the approx 5 μm layer sampled by XRD to Ti₂N and TiN—the diffractograms do not reveal any peaks of titanium. Accordingly, the dissociation pressure of CrN at 860° C. is too high.

3. Si/Si₃N₄at 860° C. yields basically the same results as Cr/CrN.

4. Nitriding with V/VN at 860° C. does not result in formation of TiN, only Ti₂N, and the magnitude of the Ti₂N peak is much weaker than in the case of CrN—Cr and Si₃N₄—Si. The titanium peaks appear to be slightly shifted versus those in the TiN—Ti diffractogram towards larger Bragg angles, thus indicating a lattice contraction instead of the lattice expansion expected for nitrogen uptake. The explanation seems to be that vanadium accumulates near the alloy surface, and alloying titanium with vanadium actually causes the lattice parameter to decrease.

In conclusion, TiN—Ti powder packs are not useful for nitridation, because the nitrogen activity they generate is too low. The nitrogen activity generated by the other metal-metal-nitride combinations at 860° C. however, is too high. In order to obtain appropriate nitrogen partial pressures, the temperature of the powder pack needs to be reduced but the specimen temperature should be kept at 860° C. to obtain large case depths in reasonable nitridation time. Therefore, we performed experiments with powder packs kept below 860° C., using long-ampoules and the two-furnace technique described before:

FIG. 12 shows X-ray diffractograms (XRD) demonstrating the best results we have obtained with Si₃N₄—Si powder-packs. The Ti-6Al-4V specimens from which we obtained these data were treated using powder pack temperatures of 400, 500, and 700° C. for 16 h. The diffractogram drawn with a solid thin line (“Ti5-Annealed”) is a diffractogram from untreated Ti-6Al-4V. The other three diffractograms were obtained from Ti-6Al-4V treated with Si₃N₄—Si powder packs at 860° C. in Ar for different times (72 to 120 h), while the powder pack was kept at 700° C. (in two cases the powder pack was equilibrated (pre-annealed) for 48 or 72 h). The results are as follows:

1. As expected for Ti-6Al-4V, the diffractogram of the untreated material exhibits a strong β-phase peak. The diffractograms of the nitrided specimens, however, do not show this peak. This means that nitridation effectively transforms the surface layer to the α-phase.

2. The diffractograms of all three nitrided specimens are similar and only differ in small details.

3. None of these diffractograms exhibit any evidence of nitride formation.

4. Each of these diffractogram exhibits a pronounced shift of the α-Ti peaks compared to the untreated material. The increased lattice parameters result in peak shifts indicating a nitrogen content>10 at %.¹⁹

5. The “best” result (the strongest peak shift and sharpest peaks, indicating the largest case depth) were obtained for the specimen that was treated for 120 h after equilibrating the powder for 48 h.

When first experimenting with the long ampoules, the formation of titanium oxides was observed on the specimen surface. It is thought that the oxygen supply for the formation of the oxides originated from the gas layer adsorbed on the inner wall of the ampoule. The reason why this did not happen with the short ampoules might be explained by the fact that when sealing the ampoules at one end, the short ampoules heat up to temperatures above the desorption temperature of water. The long ampoules, in contrast, remain cold in the regions far away from the heated end, enabling adsorbed H₂O (and perhaps other gases) to stay in the ampoule during sealing. This problem can effectively circumvented by heating the long ampoules with an ohmic resistor (heating tape) wrapped around the ampoule prior to sealing.

In addition to the XRD studies, X-ray photoelectron spectrometry (XPS) were employed, (also known as ESCA, electron spectroscopy for chemical analysis) to investigate the surface composition of the nitrided specimens. FIG. 13 which includes FIG. 13a and FIG. 13b presents two XPS spectra. Spectrum (a) seen in FIG. 13a was obtained from as-received Ti-6Al-4V, while spectrum (b) was obtained from a nitrided Ti-6Al-4V specimen after polishing 3 μm of material off the surface. Owing to the strong affinity of titanium for oxygen and carbon, the XPS spectrum (a) reveals substantial oxygen and carbon peaks. Spectrum (b) seen in FIG. 13b reveals a carbon signal of about the same magnitude as in spectrum (a) and an oxygen peak with a somewhat higher magnitude.

While the oxygen and carbon peaks are similar in FIGS. 13a and 13b, the nitrogen signals are distinctly different. FIG. 13a displays practically no nitrogen, while a clear and strong nitrogen peak is observed in FIG. 13b. Quantification of this signal yielded a nitrogen content of 13.3 at %, in good agreement with the results obtained by XRD.

These results confirm that we have identified processing conditions that generate a nitride-free solid solution of nitrogen at the surface of Ti-6Al-4V alloys. Recalling that the XRD data from the as-treated surface did not reveal titanium nitrides, implying that the surface concentration of nitrogen is below 14 at %, a nitrogen concentration of about 10 at % 3 μm below the surface suggests a case depth on the order of 10 μm.

After nitridation, the specimens obtained with a Si₃N₄—Si powder pack exhibited a significant increase in surface hardness. The surface hardness of the freshly exposed surface was measured by Vickers indentation with different loads. FIG. 14 presents a few scanning electron microscopy images of such Vickers hardness indents. The scatter of the hardness data as well as the apparent surface hardness increases with decreasing load. This effect can be explained by the fact that indents with large loads penetrate more deeply than the thickness of the hardened layer, which we estimate to be about 10 μm, as mentioned before. Thus, at higher loads the indenter penetrates into the softer material below it.

FIG. 15 presents the average hardness we have measured for the untreated material after the three different nitridation treatments of FIG. 12 with different indenter loads. Apart from the effect of the indenter load, the data reveals a substantial increase of the surface hardness. While the hardness of the untreated material was only about 375 HV, the average hardness of the best specimen, measured with a load of 50 g, amounts to nearly 1000 HV50. This result is close to the hardness one would expect from FIG. 2 for a nitrogen content of about 10 at %. The results clearly show that powder-pack nitridation can more than double the surface hardness of Ti-6Al-4V by a factor of 2.

While we have established the processing conditions for effective surface hardening by Si₃N₄—Si powder packs with success, there is substantial experimental evidence indicating that the nitrogen partial pressure generated by these packs is much higher than expected from thermodynamic data. Although we have not clearly identified the source of the additional nitrogen, additional experiments we have carried out have shown that the source resides in the Si₃N₄or Si powder itself, and not anywhere else in the system, and that the excess nitrogen is given off only for a limited time while or shortly after the powder pack is heated up to the processing temperature.

In order to avoid this apparently beneficial but somewhat uncontrolled effect, we have incorporated a procedure to remove excess nitrogen from the system before the specimen is brought up to temperature (FIG. 16): In the first step, elemental Ti is heated up in the center of the ampoule to activate it as a getter for undesired gases. Subsequently, in the second step, the powder pack is brought up to temperature. During this phase, the Ti getters any initial excess nitrogen and other gases that may be released by the powder pack. Eventually, the Ti is cooled down to room temperature, such that it becomes inactive and cannot release the adsorbed gases. In step 3, finally, the specimen is brought up to temperature and nitrided.

Applying this procedure to Si₃N₄—Si powder packs, we found that the nitrogen activity obtained under a “clean” conditions is much too small for effective nitridation of Ti-6Al-4V. In order to increase the nitrogen activity, we returned to CrN—Cr powder packs. While these initially appeared to produce too much nitrogen, they turned out to work ideally after applying the cleaning procedure of FIG. 16.

FIG. 17 presents a set of X-ray diffractograms obtained from Ti-6Al-4V nitrided for 24 h at 860° C. in clean CrN—Cr power packs held at various temperatures. The diffractogram obtained with the pack at 800° C. exhibits a strong peak shift but also a well-developed nitride peak at 39.2°, indicating that the temperature of the powder pack was too high. The diffractogram obtained with the pack at 700° C., in contrast, does not reveal nitride formation but hardly shows any peak shift compared to the reference diffractogram obtained from annealed Ti-6Al-4V. The best result was obtained with the pack at 775° C.: a pronounced peak shift, but no apparent nitride formation. The inset in FIG. 17 indicates the corresponding Vickers hardness obtained with a load of 50 g.

The most successful nitridation experiment was conducted as follows. The CrN—Cr powder pack was “cleaned” by heating at 675° C. for 24 hours while the Ti getter was kept at 800° C. and the work piece was at room temperature. Nitridation commenced by heating the specimen to 860° C. for 72 hours while the getter was at room temperature and the powder pack was kept at 675° C. for 24 hours, then at 650° C. for 24 hours and finally at 625° C. for 24 hours. Each change in temperature causes a corresponding change in nitrogen partial pressure. The steps included heating the work piece; exposing a surface of the work piece to a nitrogen gas having a first pressure that is determined by a second temperature to which a first powder of metal nitride and a second powder of the corresponding metal forming the metal nitride are exposed; and maintaining the work piece in the nitrogen gas for a period of time so that nitrogen was absorbed onto the surface of the work piece and diffused into the work piece for a desired distance while preventing the formation of nitrides on the surface of the work piece and within the work piece. In addition, the surface of the work piece was exposed to a nitrogen gas whose pressure was at a second pressure and then a third pressure different than the first pressure. The X-ray data, (FIG. 18) shows that the nitrided sample surface is a single phase α-Ti alloy; the lattice expansion indicates a nitrogen concentration of ˜8 at %; the inset show that the hardness has been increased by a factor of ˜2 compared with the untreated alloy. The tensile stress strain curves in FIG. 19 show that nitridation has not seriously affected the yield stress or the flow stress, but has caused a modest decrease in ductility (almost certainly because the surface is now single phase α-Ti, rather than the α/β mixture of the bulk of the material). A SEM micrograph of the tensile surface near the fractured region (not shown) showed slip traces and only a modicum of microcracking, similar to the tensile surface of the non-nitrided material.

In order to study the effect of nitridation on the microstructure of Ti-6Al-4V (TEM) transmission electron microscopy has been employed. A conventional TEM bright-field image of Ti-6Al-4V after nitridation with a Si₃N₄powder pack and subsequent indentation with a Vickers hardness indenter was shown in FIG. 20.

The disclosed invention for nitriding titanium and titanium alloys without nitride formation, i.e., gas-phase nitridation with “kinetic control of the nitrogen activity,” has very broad applications. With this method, case hardening by homogeneous solid solutions of nitrogen can be achieved for titanium or any titanium alloy. A technique and concept has been disclosed herein to generate appropriate nitrogen partial pressures has been successfully demonstrated. Also an example of the technique has been described wherein encapsulation and tempering of the workpiece in an evacuated container with powder packs of a suitable metal-metal-nitride combination, e.g., Cr—CrN, has shown to be operable.

Although the invention has been shown and described with respect to a certain preferred embodiment or embodiments, certain equivalent alterations and modifications will occur to others skilled in the art upon the reading and understanding of this specification and the annexed drawings. In particular regard to the various functions performed by the above described method and components (assemblies, devices, circuits, etc.) the terms (including a reference to a “means”) used to describe such components are intended to correspond, unless otherwise indicated, to any component which performs the specified function of the described component (i.e., that is functionally equivalent), even though not structurally equivalent to the disclosed structure which performs the function in the herein illustrated exemplary embodiments of the invention. In addition, while a particular feature of the invention may have been disclosed with respect to only one of several embodiments, such feature may be combined with one or more features of the other embodiments as may be desired and advantageous for any given or particular application.

Surface hardening of Ti alloys by gas-phase nitridation: kinetic control of the nitrogen activity

Information

Publication Number

Date Filed

Date Published

Inventors

CPC

US Classifications

International Classifications

Abstract

Description

Claims

CROSS-REFERENCE TO RELATED APPLICATIONS

Provisional Applications (1)