Hot isostatic pressing (HIP) of MIM

Table of Contents

• 9.1 Introduction

• 9.2 The HIP process

• 9.3 Benefits of the HIP process

• 9.4 Issues with the HIP process

• 9.5 Example HIP processes conditions

• 9.6 Conclusion

9.1 Introduction

Hot isostatic pressing (HIP) was invented at Battelle Memorial Institute in the 1950s for the diffusion bonding of nuclear reactor components and was soon found to be a great tool for consolidating powders and removing porosity in hardmetals. HIP has evolved into an industrially accepted process for densifying metals and alloy to increase their mechanical properties and to increase their polishing and plating ability. It is primarily used to remove internal voids. Most metal injection molding (MIM) producers outsource this operation to an external vendor that specializes in HIP. The process is simple and consists of high temperature and high pressure on components. For that purpose, the HIP equipment is specially designed to sustain these conditions and can be very expensive. During HIP, the high temperature achieved makes the material soft enough to deform and the high pressure provides the force to compress the internal pores. Pores are eliminated by a creep and diffusion mechanism. The yield strength (YS) of the material decreases due to the temperature and the argon gas has the atomic size to apply pressure without diffusing through the metal. Creep mechanisms that are active during the initial stages of HIP consist of Nabarro-Herring creep (diffusion through grain interiors), Coble creep (grain boundary creep), and dislocation creep. Final-stage HIP consists of diffusion bonding of the closed pore walls to each other. MIM components are well suited for HIP since the as-sintered density is sufficiently high to isolate the internal voids from the outside (not interconnected) and thus be compressible. If the MIM component has interconnected pores that are exposed, the argon compression gas will simply fill these pores and not compress the pores. In bulk metal manufacturing by HIP of powders, the powders are put into a metal can, the can is evacuated, and welded shut so that the gas and heat can compact and bond the powder together. The can is required to allow compressibility of the powders. Fortunately, MIM typically sinters to densities above 95% and in most cases 98% plus, which is well above the minimum density required to densify without a can. The minimum density for HIP of noncanned materials is in the 92%-94% range. Table 9.1 provides a minimum density guideline to allow canless HIP of multiple metals and alloys. MIM components can be practically tested for its ability to HIP without a can by writing on the surface using a permanent marker. If the marker shows clearly and does not wick into the part, it can be HIPed for greater density. If it is wicked into the part, it cannot be HIPed for greater density.

Table 9.1 Minimum density to allow HIP of MIM components

9.2 The HIP process

The HIP process utilizes hot compressed inert gas to apply pressure to the work. In the case of MIM, HIP temperatures are typically 100-200℃ less than the sintering temperature and pressures are usually in the range of 15,000-20,000 psi (105-140MPa). The preferred gas is argon since it has a large atomic size; however, nitrogen can be used, but may not be as effective. Fig. 9.1 is a schematic of the process. The process is a batch process that typically lasts 4-10 h "door to door." It has the following sequence of steps:

1. Load work in chamber and close;

2. Evacuate air and fill with inert gas;

3. Pressurize and heat at the same time;

4. Cool and depressurize at the same time;

5. Vent; and

6. Unload work.

This sequence represents the stand alone HIP process and is commonly used for canned billets investment cast, and MIM components. In the carbide industry, a process called sinter-HIP or pressure sintering is used to densify the carbide/cobalt powder-based components. The sintering and HIP is done in one step to reduce the overall time and cost. The pressure is typically 1.5-10MPa which is much less than the standard HIP pressure, however sufficient to remove the last bit of porosity of the carbide.

Fig. 9.1 Schematic of HIP process.

9.3 Benefits of the HIP process

The primary reason to employ HIP is to improve the mechanical properties of metals by eliminating porosity. As previously stated, MIM is the ideal candidate for HIP. The innate sintering density obtained by MIM is sufficiently high to isolate all but a small amount of surface porosity, thus densification can be done without the use of cans, which is usual practice for making HIPed billets. HIP has multiple attributes which enhance the final product. It densifies the part, thus giving it greater properties, more uniform dimensions, better surface finish, and also less chance for exposure of pores during polishing. Fig. 9.2 shows a microstructure of 316L before and after HIP, notice elimination of pores and the grain growth to the right. Grain growth has been observed for 17-4pH SS and F2886 of MIM parts. LaGoy reported a five to six times increase and Sago reported three times increase in grain size.

Fig. 9.2 MIM 316L sintered at 1350℃ before (left) and after HIP at 105MPa and 1100℃ (right). Notice the grain growth after HIP.

HIP will densify any pore that is not connected to the outer surface. To be susceptible to closure, a pore needs to be closed off from the outside. Fig. 9.3 is an illustration of a pore that will not densify and of one that will densify.

Fig. 9.3 Illustration of pores: that (B) will and (A) will not densify.

Another benefit of HIP after MIM is to reduce the dimensional variability of components. Since the components are densified to maximum density possible with HIP, the overall dimensional variability will decrease from one component to another following HIP. A common sintering technique is to sinter multiple components in a batch furnace. Envision a batch furnace load that has a variation in temperature profile from one corner to the center of the hot zone. If the parts in the center densify to 98% density and the parts in the corner densify to 96% density, a dimensional variation exists between these two parts. Now envision taking both of these parts and putting them into a HIP cycle. After the HIP, both parts will have a density near 100%; therefore, the final dimensions will more closely match each other.

HIP of MIM will increase the MIM mechanical properties. A rule of thumb in metal component powder processing is as the density increases, the properties improve. Hardness, YS, and ultimate tensile strength (UTS) will all show some improvement but the greatest improvement will be for the dynamic properties such as elongation, fatigue, and impact strength. These improve dramatically since they are susceptible to microstructural defects. Table 9.2 reports some of the properties of a few MIM materials before and after HIP.

Table 9.2 Example MIM and MIM/HIP mechanical properties

In addition to increasing properties, the HIP process also improves the ability for a surface to be polished. Polishing is a method by which material is being removed from the surface. Material can be removed mechanically or electrically (electropolish). The removal of material continually exposes new material to the surface. If there is a pore just subsurface, polishing will expose it and will result in an uneven surface. Areas of pore exposure are subject to entrapment of polish media/solvent or other contaminates and can result in poor surface quality and potential contamination. If the component is HIPed prior to polishing, no subsurface pores will be available for exposure and the surface integrity will be vastly improved and the cleanliness of the surface will be enhanced. In non-HIPed samples, exposed pores can hold contamination which can leave defects during plating and harbor bacteria in medical applications.

Weldability of the MIM compact is also improved with HIP. Alloys with porosity weld poorly, therefore, the removal of porosity by HIP can improve welds.

9.4 Issues with the HIP process

HIP brings many great attributes to the final property of materials; however, there are concerns about some of the practical negatives that the process can produce. These are distortion, potential surface contamination, and variability of distortion and contamination from lot to lot.

As discussed before, HIP will densify MIM parts to near full density. In MIM, it is possible that a component has density gradients by virtue of earlier processing. For example, a component can have a density gradient within its shape that is dependent on the molding gate location. Near the gate, the component has highly packed particles and away from the gate, where the molding pressures are less, the component has loosely packed particles. During sintering, the area close to the gate will shrink less than the area furthest from the gate, if this type of gradient exists. This can be further enhanced during HIP with the less dense area shrinking more than the greater dense area, resulting in distortion and anisotropic shrinkage of the component. Additionally, anisotropic shrinkage and distortion can occur by virtue of heat gradients within the HIP unit itself.

Distortion can also occur due to cooling gradients in the HIP chamber. If the part has thin and thick sections, the thinner sections will cool faster than the thicker sections which can result in distortion. This phenomenon is also observed in heat treatment. One method to mitigate this is to control cooling of the work vessel. Although not always practical with a "shared load" method of commercial HIP processors, it could be practical if one purchases an entire run at the HIP processor. This practice can offer more control over distortion in certain component, but it could be expensive.

Components made by MIM respond well to the HIP process because little surface porosity exists and the pores resulting from the MIM process are typically very small. Unlike investment casting, no dimple or depression is seen on properly fabricated MIM components after HIP. However, sometime a slight deformation near an existing pore can be seen. If a depression is seen after HIP, the pore that causes this depression is not typically from the MIM small pores, but rather comes from a larger defect, possibly from poor molding conditions. One should review the molding conditions and perhaps investigate that area by sectioning the part prior to HIP to determine if the molding process conditions were the cause of the pore defect.

Another practical concern with HIP is the potential for contamination on the surface of components during the HIP operation. HIP vendors run multiple alloys in their HIP vessel, as such, the work can get contamination on the surface. Green and brown contaminations on the surface of components after HIP operation have been seen by the author. An analysis of this shows it to be chromium and silicon. This has been mitigated by using tool wrap foil to surround the parts to prevent this contamination or to minimize it. Vacuum solution annealing to vaporize the contaminant from the surface can also be used. Some have claimed that reheating a component at ambient pressure or vacuum will reopen the pores. Although occasionally reported in literature, it has not been observed by the author. Pores close during the HIP process by virtue of vacancy diffusion from the pore to outside of the component and the walls of the pore diffusion bond to each other during the final stage of the HIP process; therefore, the diffusion bonded pore walls have become grain boundaries and are unlikely to reopen.

9.5 Example HIP processes conditions

Typical HIP process conditions are limited. Most vendors have standard runs where the variables are temperature, pressure, and time. They set their runs at the same pressure, and varying time and temperature. Table 9.3 presents a few standard HIP conditions that are used for MIM components.

Table 9.3 Typical HIP cycles that can be used for different alloys and metals

9.6 Conclusion

MIM is an ideal candidate for HIP by virtue of the resulting post-sintering closed pore condition. MIM Components that are HIPed show improved mechanical properties increasing density up to full density, and significant grain growth. Dimensional variation is also decreased by homogenization of the potential variation seen in batch MIM sintering processes. Additional benefits include improved polishing and weldability. Most HIP is not performed by MIM manufacturers, thus, HIP is performed in a mixed workload of different geometries and in some cases alloys.