Production Evaluation of Coatings
and Surface Treatments for Die Casting Dies.

By Sandhya Gopal and Rajiv Shivpuri, Department of Industrial, Welding and Systems Enigneering, The Ohio State University
(Reprinted with permission from Die Casting Engineer, March/April 2000 issue.)

Interaction between the die surface and cast metal leads to both mechanical and chemical degradation of the surface. The mechanical degradation is caused by the high speed impingement by the melt particles of the die surface near gates. This type of die wear is called washout. The chemical affinity of the cast metal atoms for the die surface leads to the formation of intermetallic layers at the interface. This phenomenon is called soldering. Production experience has shown that oxide layers and other protective layers on the die surface successfully retard the onset of soldering and washout of these surfaces. There are many surface engineering techniques and a wide range of coatings available in the market which are potential candidates for soldering and washout prevention. The quest is to identify suitable candidates to combat the harsh die casting conditions. As part of this quest, production evaluation campaigns are being held at die casting companies with varying matrix of die wear problems. These production campaigns will help pinpoint the best surface engineering technique and composition for maximum die life increases in a predetermined production campaign. The first production campaign was held at Pace Industries in Monroe City, Mo. In this campaign, the use of coatings (chromium carbide, chromium nitride and vanadium carbide) significantly reduced the soldering tendency and the soldering-related downtime was totally eliminated for the production run of 180,000 shots. The results of this campaign were reported in an earlier paper published in the 1997 NADCA Transactions. This paper contains details of the second campaign that was run at Premier Tool in September-December 1998.

Premier Tool & Die Cast Corporation in Bemen Springs, MI, was selected as the ß-test site for the production evaluation campaign presented in this paper. The objective of the production evaluation campaign at Premier was to evaluate promising coating compositions and application techniques for their efficacy in preventing the occurrence of soldering on the core pins in the two 16 cavity dies for suspension mounts. This campaign was designed to involve both surface treatments and surface coatings.

Production Evaluation/Process Conditions

Premier offered a challenging testing environment in its high volume 16 cavity die. The die was used to produce automobile suspension mounts.The problems associated with the 16 cavity die were not solely soldering-related. There were issues regarding cooling, gating and load bearing capacity of the ejector pins too. Parts were sticking to the ejector half. The cores were snapping at the head. Soldering on the cores was severe especially near the parting line. The cores were removed 1-2 times every shift for polishing. 1-4 cores broke per run. The die had to be removed every 5,000 shots for die maintenance.

Premier addressed some of the above problems by incorporating design changes. The dimensions of the core were modified to eliminate the breaking of the head. Heating and cooling lines were provided within each core to reduce the occurrence of potential hot spots that would promote soldering. The dimensions and position of the ejector pins were modified to facilitate ejection.

The production evaluation was conducted on two 16 cavity-dies. The process conditions for both the dies were same. The geometry of the core on the ejector half was different from the geometry of the core on the cover half for both the dies. The only difference between the cores belonging to the dies was in the geometry of the core. Die 1 had oval cores in the cores and Die 2 had round cores. The cores on the cover half of Die 2 were DME pins. The cores on the ejector half of both the dies, and the cores on the cover half of Die 1were fitted with internal cooling. A 1/8th-inch copper pin was used to extract the heat and a 1/8th bubbler with high flow cascade was used to cool the core.

The process conditions for the production evaluation were:

Machine size:	600 Ton
Number of die cavities:	16
Scheduled Typical Number of shots/day:	1300 shots
Alloy:	A380
Temperature of the melt:	1250°F
Average Temperature of the die surface:	550°F
Cycle time:	45 seconds
Spray Lubricant:	Chemtrend 2258
Dilution ratio:	60:1

Coating/surface treatment selection criteria

Premier, in its 16 cavity die, offered many potential sites for severe soldering and washout problems. The cores on the cover half of Die 2 did not possess a cooling line and were highly prone to soldering. The geometry of the core pins was complicated with non-uniform dimensions that made efficient heat transfer tough. Hence an effective coating or surface treatment candidate, for combating the conditions at Premier, needed to possess the following properties:

• Sound adhesion to the substrate
• Sufficient hardness and toughness
• Low chemical affinity for and low solubility in molten aluminum
• Superior oxidation resistance
• Good thermal and impact shock resistance
• High thermal conductivity to dissipate heat from the interface quickly
• Compatible Coefficient of thermal expansion with the
  substrate

A potential surface engineering technique should be able to deposit the coating uniformly in spite of the complexity of the geometry. This surface engineering technique could either be a coating deposition process or surface treatment process.

• If material with desired properties is added to the surface, then the process is called a Coating Deposition Process
• If the chemistry and/or microstructure of the substrate of the base material is altered, then the process is called a Surface Treatment Process

Selected Candidates

Physical Vapor Deposition (low deposition temperature) and Thermo-reactive diffusion process (high temperature deposition process) were identified as potential coating techniques. Nitriding and Carburizing were identified as potential surface treatment techniques. A duplex treatment was also included in the design to compare the performance of a surface textured PVD coating and a smooth PVD coating. The micro texturing of the substrate was provided by shot micro-peening. Three coatings, CrNx, CrxCy and BxC, were the selected physical vapor deposition (PVD) coatings and VC was the selected thermo-reactive diffusion (TRD) coating. CrNx (PVD) on shot peened substrate was the selected duplex treatment. Ferritic Nitrocarburizing and ion nitriding were the selected surface treatments. Table 1 includes relevant information about these selected candidates.

Technique	Suppliers	Coating / Surface Treatment	Coating Thickness	Hardness
Physical Vapor Deposition	Multi-Arc	CrNx	6-8 µm	2500 HV
	Balzers	CrxCy	10 µm	1850 HV
	Diamond Black	B2C	2 µm	900 HV
Thermo-Reactive Deposition	TD Center	VC	7-10 µm	3000 HV
Surface Treatment	Advanced Heat Treatment	Ultraglow (Ion Nitriding -1)	0.15 - .020 mm (case depth)	697 - 1070 HV
	Sun Steel Treating	Ion Wear (Ion Nitriding -20)	0.08 - 0.13 mm (case depth)	746 HV
	Dynamic Metal	Ferro-Nitro-Carburizing	0.13 - 0.25 mm (case depth)
Duplex	Badger Metal	Shot Peening + CrN	6-8 µm	2500 HV
Substrate	Thyssen	H13	—	46 - 47 HRc

Table 1 Selected Candidates and their properties Physical Vapor Deposition

Physical Vapor Deposition (PVD) processes deposit coatings on a substrate atomistically, i.e. atom by atom. The material to be deposited is transported in the form of vapor, either through a plasma or vacuum, to the substrate on which the vapor condenses. The source for the vapor could either be thermal or non-thermal. These processes can deposit both single elements and compounds as coatings. The thickness of the coatings can vary from a couple of nanometers to a few millimeters. All PVD processes are line of sight processes. Usually, PVD coatings possess columnar structure which is not as good as equiaxed structure in combating liquid metal corrosion. The columnar grains provide a pathway for the molten alloy to diffuse through. Physical Vapor Deposition is classified into three types: evaporation, sputter deposition and ion plating. The production campaign had candidate coatings applied by either evaporation process or by sputter deposition.

• Arc Evaporation: Vacuum evaporation takes places at gas pressure ranges of 10^-5 Torr to 10^-9 Torr. The coating material is in an electrically neutral state and is expelled from the surface of the source at thermal energies typically from 0.1 to 0.3 eV The substrate is preheated to elevated temperatures (200 to 1600°C) for dense and equiaxed grain morphology.
  1. CrNx: This coating was provided by Multi-Arc Inc which utilized the arc evaporation process to deposit the coating. In this process, a vapor plasma is generated by striking an arc between the solid cathode (target) and the arc source. The arc melts a small area (10 micrometers) of the cathode surface generating metal droplets (Cr), ions and large volume of free electrons. This vapor is highly ionized (up to 80%) and arrives at the substrate with high energies (50 eV). The substrate temperatures are in the range 200 to 550°C. Nitrogen gas is inducted in the vacuum chamber to create nitrides. Process times are of the order of 4 hours for a coating thickness of 6 microns.
• Sputter Deposition: In this process, the substrate is deposited with particles vaporized from a surface, which is called the sputtering target. It is a non-thermal vaporization process where the coating material is dislodged from the surface of the target by momentum transfer from energetic particles which bombard the surface. The substrate is positioned in front of the target so as to intercept the flux of sputtered atoms. Sputter deposition can be performed in a vacuum or low pressure gas (<5 mTorr). Sputter deposition can also be deposited at higher gas pressures (5-30 mTorr).
  1. CrxCy: This coating was provided by Balzers Tool Coating Inc., which applied it by the e-beam sputtering process. In the e-beam evaporation process, the surface of the workpiece is bombarded with noble gas ions in order to remove contaminants and to sputter off some substrate material. These substrate atoms then condense with the coating element (Cr) which is then evaporated in the second stage. Reactive gas (carbon) is then introduced into the chamber, which combines with the chromium ions on the surface of the workpiece to form hard CrxCy coatings.
  2. BxC: This coating was provided by Diamond Black Inc. It was applied by magnetron sputtering process, which is performed at low temperature (250°F). The coating was vacuum sputtered to a thickness of 0.00008" or 2 microns at 250°F.

Thermo-Reactive Diffusion Process

In the TRD process, the carbon and the nitrogen in the steel substrate diffuse into a deposited layer with carbide-forming or nitride-forming elements such as vanadium, niobium, tantalum, chromium, molybdenum or tungsten. The diffused carbon or nitrogen reacts with the carbide and nitride forming elements in the deposited coating so as to form a dense and metallurgically bonded carbide or nitride coating at the substrate surface. The possibility of distortion is present with this high temperature process. Dimensional changes due to the high phase transformations in the heat treatment of the base steel and the formation of the carbide layer are a good possibility. The coatings formed have a fine and nonporous composition. Though the diffusion layer is thin, it is very dense and shares a sound metallurgical bond with the substrate.

VC: This coating was supplied by the TD Center using the TRD process. The high temperature salt bath TRD process was performed in a molten borax bath at 850 to 1050°C (1560 to 1920°F). Immersion time ranges from 0.5 to 10 hours to obtain an optimum carbide layer thickness of 7 -10µm.

Duplex Treatment

Metalife+ CrNx; The duplex coating was formed by first shot peening (micro-shot) the substrate and then coating it with CrNx (Arc Evaporation PVD). The CrNx coating was applied by the arc evaporation technique by Multi-Arc. The shot peening treatment was supplied by Badger Metal. Badger Metal Technologies applied a patented micro-peening treatment (Metalife) on the core pins and the DME pins. The treatments for die casting are categorized by "T" processes (T10, T21, T41, T61 and the newer T71). The surface of the pin is impacted with special media, a temporary plastic flow of the metal (with penetration depths of 0.010 to 0.015 in. on 44 to 48 HRC surface) results in generation of compressive residual stresses inside the peened surface layer.

Surface Treatments

1. Ferritic Nitrocarburizing: Ferritic Nitrocarburizing processes are thermochemical processes, which involve the simultaneous diffusion of both nitrogen and carbon to the surface of ferrous materials at temperatures completely in the ferriric phase field. The primary objective of such treatment is to improve the anti-scuffing characteristics of ferrous engineering components by producing a "compound layer" on the surface, which has good tribological properties. A single phase epsilon carbonitride compound layer is produced supported by nitrogen rich subsurface diffusion zone. Dynamic Metal Treating Inc. uses fluid bed (salt bath) ferriric-nitrocarburizing at (600°F to 1000°F) and steam blueing. The ferriric carburizing process takes between 4 to 12 hours. Steam blueing is achieved by sealing in a tempering furnace at 350°C to 370°C and introducing steam. This process creates a blue black surface finish due to the formation of a tight blue oxide layer. A surface hardness higher than 70 HRc is achieved. The growth in dimensions, due to the process are, of the order of 0.0001 to 0.0002" per side. Typical compound zone depths of 0.0005 to 0.001" and case depths of 0.005" are achieved. Dynablue 10B was used in the tests.
2. Ion Nitriding: Ion nitriding is an extension of the conventional nitriding process using the plasma-discharge physics. In vacuum, high-voltage electrical energy is used to form plasma through which nitrogen ions are accelerated to impinge on the workpiece. This ion bombardment not only cleans the workpiece surface but also heats up the surface and provides active nitrogen for nitriding. Two different companies supplied the ion nitriding treatments for the campaign at Premier.

   a) Ultraglow: Ultraglow process, applied by Advanced Heat Treat, is basically ion nitriding process with treating conditions optimized to reduce the white gamma prime compound zone on nitrided surface. The nitriding parameters were adjusted to yield an average surface hardness of 90 to 94 HR15N on the core pins for Premier Tool and Die Cast and 90 to 91 on the DMG pins. The case depths are maintained from 0.006 to 0.008 in.
   
   b) Ion Wear: lon Wear treatment was provided by Sun Steel. This process involves a combination of ion nitriding at 400 to 565°C and steam treating (oxidation). This diffusion treatment creates a multi-layer composed of complex oxy-carbo-nitrides up to 0.0004" (10 micrometer) and case depths up to 0.025" (0.6 mm). For the Premier test the treatments had a surface hardness of 800-850 HV1, a compound layer of 0.0002 in (5 microns) thick and case depth of 0.005 to 0.0009 in (75 to 125 microns)

Results from Production

Downtime Statistics - The H13 core pins were polished once every 5000 shots approximately to remove the soldering. The frequency of polishing for removing the solder build up reduced since the use of coated and surface treated core pins. The polishing statistics with the coated and surface treated cores is listed below and included graphically in Figure 1.

Fig. 1 Polishing statistics with the coated and surface treated cores.

For Die # 1(Oval Cores):
1. The first polishing was done after 7330 shots.
2. The second polishing was one after 14,932 shots
3. The third polishing was after 32,369 shots

For Die # 2 (Round Cores):
1.The first polishing was after 12,983 shots
2. The second polishing was after 22,181shots
3.The third polishing was after 29,116 shots

Visual Inspection of the Cores

When there were problems related to ejection of the part, production was stopped and the cores were visually inspected after polishing them with emery paper. The operator then decided whether a particular core pin was fit to be used in production again. If the solder build-up was severe that effective removal by polishing was not possible, the cores were not used in production again. Results of the visual inspection are included in Figure 2. In Figure 2, for 0-7330 shots, A represents no soldering, B means marginal amount of soldering and C represents significant amount of soldering. For 7330-14932 shots, C means the core was removed from service or was on the verge of being removed from service due to severe soldering. For 14932-32269 shots, A means the core is still running, B means the solder build-up on the core may or may not clean up and C means the core has failed.

Discussion

A generalized conclusion could not be drawn about the efficacy of the various coating and surface treatments in reducing soldering from the production results. BxC, applied by magnetron sputtering, failed at 14,932 shots while CrxCy applied by E-Beam sputtering, is still running after 32,269 shots. VC, applied by TD process, is also still running after 32,269 shots. Among the ion nitriding, Ultraglow performed better than lon wear. The increase in life time of the H13 core pin does not appear to be significant but it should be borne in mind that the H13 cavity die possessed severe soldering conditions.

The performance evaluation of the coatings has been done only for the ejector half of Die 1. Due to severe sticking in the cover half, all the coated cores except the one coated with VC were replaced with cores coated with Ferritic Nitrocarburising (FNC). FNC was chosen due to prior experience with it in production.

Fig. 2 Graphical Representation of the visual inspection results.

From Figure 2, the following deductions can be made about carbide and nitride coatings.

Carbides: Carbide coatings, in general, have performed better than the other coating compounds. A comparison across techniques was done to evaluate the carbides further.

• Surface Treatments: FNC is the only candidate that belongs to this category. It was applied by low temperature nitrocarburizing process followed by steam blueing. According to the visual inspection report, core 14 was removed from service after 14,932 shots due to excessive soldering. But core 13 had survived 32,269 shots. The oxide layer formed on this core due to steam blueing could have protected the surface nitrogen from diffusing inwards and the carbon from getting oxidized. Beneficial effect of an oxide layer produced by steam has been reported by Norstrom. The protective oxide film that forms isolates the metal from its environment and it acts as a passive layer.
• PVD: Both CrxCy and B4C belong to this category. Both are low temperature sputter deposited coatings.
  1. CrxCy was applied by e-beam sputtering. Both the CrxCy coatings had survived 32,269 shots. CrxCy possesses excellent mechanical properties. Various hardness values have been reported for CrxCy but all of them are between 1000-1650 kg/mm² It possesses excellent oxidation resistance and has an oxidizing temperature which is higher than 800°C. The solubility of Chromium in aluminum is low too. The performance of CrxCy could have been effected by die nature of its residual stresses. Sputter deposited CrxCy has a compressive residual stress in the order of 1 Gpa [1]. A compressive residual stress plays an important role in preventing the propagation of cracks.
  2. B4C was applied by magnetron sputtering. Both the B4C coatings were removed from service due to excessive soldering at 14,932 shots. The minimal thickness of the B4C coating, 2 microns, could have contributed to its failure. Though B4C has a high hardness of 4700 kg/mm² [1], it has very poor toughness. To combat erosive wear, toughness of the protective barrier is important.

• Fig. 3 VC coated oval core on cover half of Die 1- shots seen 32,269 shots (Scale 1:4)

TRD: This high temperature diffusion process was used to apply the VC coating. Core 5 was removed from service at 14,932 shots for evaluation of two areas of solder. Core 6 survived 32,269 shots. VC coatings have a fine and nonporous composition which resist the diffusion of aluminum through them. Though the diffusion layer is thin, it is very dense and shares a sound metallurgical bond with the substrate. The metallurgical bond, because of its excellent adhesion strength, prevents the spalling of the coating. It has high hardness of 3000 HV. It exhibits a thermal expansion mismatch with H13 that contributes to thermal stresses.

• Nitrides: A comparison across techniques was done to evaluate the nitrides.
• Surface Treatments: Ion nitriding is the only surface treatment that falls in this category. From table 1, it can be noted that two different companies supplied the ion nitrided core pins.
• Ultraglow: Core 3, treated with IN 1, chipped in the first run and was removed from service after 7,330 shots. Core 4, treated with IN 1, failed at 32,269 shots. Ultraglow, which utilizes the basic ion Nitriding principles, is performed at room temperature. Hence when the ion nitrided core pin was placed at high die casting temperature, the nitrogen ions migrated towards the higher temperature. This migration is due to thermal
  diffusion and rendered the treatment ineffective after a large number of shots.

• Fig. 4 Ion Nitrided (2) coated oval core on ejector half of Die 1 - failed at 21,679 shots (Scale 1:4)

  Fig. 5 CrNx coated oval core on ejector half of Die 1 - removed after 24,939 shots (Scale 1:4)

  Fig. 6 Metalife treated fitted oval core pin on ejector half of Die 1 (2000-3000 shots)

lon wear: The performance of IN2 was different than that of IN1. IN2 uses a combination of ion nitriding and steam treating (oxidation). Core 7 treated with IN2 failed at 21,679 shots. Core 8, treated with IN2, was removed from service at 32,269 shots. The failure of this core could have been caused by the oxide layer being penetrated by the aluminum alloy.

• PVD: CrNx is the only nitride that was applied by low-temperature arc evaporation process. Core 1 had seen 7,609 shots in total. Core 2 was not installed. The hardness of CrNx is 1424 HV and it possesses a toughness of 17N [1]. Cr has a low solubility in aluminum. CrNx is stable until 800C, after which its oxidation resistance decreases. The stability of the oxide layer at die casting temperature protects the coating from scaling. The performance of the arc evaporated CrNx could have been effected by the nature of the residual stresses. Evaporated CrNx coating has a tensile residual stress in the order of 1 Gpa [1]. Compressive residual stress plays an important role in preventing the propagation of cracks but tensile residual stress has a detrimental effect on crack propagation.
• Duplex Treatment: CrN2, which is a combination of shot peening and CrNx, (arc evaporated PVD coating), belongs to this category. Core 9 broke at 2000-3000 shots and core 10 had survived 32,269 shots. Metalife leaves a residual compressive stress on the subsurface. This compressive stress combats any stress that encourages crack propagation. Hence cracks cannot propagate through a layer of compressive stress. Even if a crack propagated through the PVD coating, it would find it difficult to propagate through metalife treatment. The textured surface of metalife treated core pins increases the efficacy of lubrication too.

Coating	Core #s	Core #s
CrNx	Core 1	16,133
	Core 2	16,133
Ion Nitriding (IN 1)	Core 3	29,116
	Core 4	12,983 - 22,181
VC	Core 5	12,983 - 22,181
	Core 6	12,983 - 22,181
Ion Nitriding (IN2)	Core 7	12,983 - 22,181
	Core 8	29,116
T-41 + CrNx	Core 9	12,983 - 22,181
	Core 10	12,983 - 22,181
CrxCy	Core 11	29,116
	Core 12	29,116
Ferritic Nitrocarburizing	Core 13	12,983 - 22,181
	Core 14	12,983 - 22,181
BxC	Core 15	12,983 - 22,181
	Core 16	12,983 - 22,181
H13	Cores 17 & 18	12, 983
	Cores 19 & 20	0

Table 2 Visual inspection data for the ejector half of the round cores

Fig. 7 H13 round core on ejector half of Die 2 - removed from service at 12,983 shots (Scale 1:4)

There is no definite data on when the coated round cores failed. Table 2 represents the visual inspection data for the round cores. In general, the coatings and surface treatments on the round cores seem to have lasted longer than those on the oval cores Some of the general observations from production can be summarized as follows:

1. Coating and surface treatments of the cores has decreased the breakage of the cores. This is primarily due to reduced soldering and ease of ejection of casting.
2. Polishing of cores for removal of solder buildup has decreased with the coated and surface treated cores. Polishing is detrimental to the PVD coatings as it removes the thin coatings. The diffusion coatings being thicker are less affected by polishing.
3. Carbides (chromium carbide, vanadium carbide and ferriric nitrocarbide), in general, have done better than the other compounds. This may be due to their greater thermal stability.
4. Vanadium and chromium carbide were thicker coatings and performed well while Born carbide, being very thin, did not perform as well.
5. Diffusion treatments being thicker performed well as they have concentration (hardness) gradients as deep as 100 microns below the surface.
6. Die casters should be very careful when using thin PVD coatings: DO NOT POLISH THE PINS. The coatings are too thin. They can he chemically cleaned instead.

Acknowledgements

The authors thank NADCA Surface Engineering Task Committee (Peter Ried/Chair) and NADCA for having supported this study financially. They are grateful to Premier Tool & Die Cast Corporation and Jeff Brenan for their cooperation and support. They express their thanks to the Coating and Surface Treatment Companies for coating the core pins often under time pressure. Their "in kind" support was critical to the success of this project.

This article, Production Evaluation of Coatings and Surface Treatments for Die Casting Dies was originally published and copyrighted in 2000 by the North American Die Casting Association (NADCA) in Die Casting Engineer, in March 2000. It is published here with the permission of NADCA and may only be republished with the permission of NADCA by contacting dce@diecasting.org. NADCA is the worldwide leader in stimulating growth and improvement in the die casting industry.

TD Center
2020 15th Street, Columbus, IN 47201
Ph: 877-832-3687 • Fax: 812-378-1591