Mar 06, 2024 Leave a message

Primary Carbide Formation in Electroslag Remelted H13 Tool Steel

Hot work die steel H13, designated 4Cr5MoSiV1 in GB/T 1299 and H13 in ASTM A681, is the workhorse of hot forging, die casting and hot extrusion tooling. Its service life is decided less by the base hardness than by the size, type and distribution of carbides that survive solidification. When H13 is remelted by electroslag remelting, the very slow solidification of a large ingot encourages segregation, and the last liquid to freeze is enriched in carbon, vanadium, chromium, molybdenum and nitrogen. That liquid ends as coarse eutectic carbide and nitride networks which downstream forging and heat treatment cannot dissolve. This article sets out how those primary phases form, how they are measured, and which melting, forging and heat treatment parameters keep them fine.

H13 Chemistry and Why Carbide Control Defines Tool Life

H13 is a 5% chromium hot work steel with molybdenum and vanadium additions that give it tempering resistance and hot hardness. The specification ranges are narrow because the balance between carbide formers and carbon determines how much primary carbide can appear at all. The typical analysis to GB/T 1299 for 4Cr5MoSiV1, consistent with AISI H13 in ASTM A681 and with ISO 4957, is given below.

Element Specified range, % Role in carbide formation
C 0.32-0.42 Supplies carbon for MC, M7C3 and M23C6 carbides
Si 0.80-1.20 Deoxidation and tempering resistance, raises activity of C
Mn 0.20-0.50 Hardenability, ties up sulphur
Cr 4.75-5.50 Forms Cr-rich M7C3 and M23C6, raises hardenability
Mo 1.10-1.75 Secondary hardening, partitions to M6C and M2C type carbides
V 0.80-1.20 Strong MC former, controls grain size, forms primary VC
P 0.030 max Segregates to last liquid, raises temper embrittlement risk
S 0.030 max Inclusion control, usually held much lower in ESR stock

Carbon at 0.4% with 5% chromium and 1% vanadium puts the steel well inside the hypereutectic region for the carbide-forming system, which means that on slow cooling a eutectic reaction is unavoidable: austenite plus carbide of the M7C3 type, plus a vanadium-rich MC component. In a small continuously cast billet the eutectic cells are small and can be broken up by hot working. In a large ESR ingot the local solidification time at the centre is measured in hours, the eutectic cells are coarse and interconnected, and forging reduction alone cannot fully disperse them.

Where Nitrides and Primary Carbides Nucleate in an ESR Ingot

Electroslag remelting produces a water-cooled copper mold filled with a molten slag pool, typically of the CaF2-Al2O3-CaO family with about 70% calcium fluoride and 30% alumina for a general purpose slag, sometimes with lime or magnesia additions for sulphur and oxygen control. A consumable electrode is fed into the slag, current passes through the resistive slag, and metal droplets form at the electrode tip and fall through the slag to a shallow metal pool that solidifies progressively upward.

Solidification is directional and slow. The local cooling rate at the ingot centre of a 500-700 mm diameter ESR ingot falls in the range of a few tenths of a degree per second or lower, compared with tens of degrees per second in a small continuous casting mold. Solute rejected by the growing solid cannot diffuse away over the distances involved, so the residual liquid enriches steadily. The sequence of precipitation follows the solubility product of each phase, and it is observed as a defined order: titanium nitride appears first, then vanadium-rich MC carbide appears very close to the end of solidification, and the chromium-rich eutectic carbide fills the last interdendritic space.

In published ESR studies of H13 the onset of TiN is reported at a solid fraction of roughly 0.82, with VC appearing close to the solidification end point. The physical meaning is straightforward: by the time the remaining liquid is only some 15-18% of the volume, the enrichment of titanium, nitrogen, vanadium and carbon in that liquid is sufficient to exceed the solubility product, and the phases precipitate directly from the liquid rather than from solid solution.

Thermodynamics of TiN and VC Formation at the Solidification Front

The controlling quantity is the activity of each component in the residual liquid, not its bulk concentration. This distinction explains an otherwise puzzling observation: at the end of solidification the vanadium and carbon concentrations in the residual liquid are already high, yet the measured carbide is not what a simple calculation assuming unit activity would predict.

The reason is that the carbide that forms is not pure vanadium carbide. Chromium, molybdenum and iron dissolve in it, so the activity of VC in the precipitate is well below unity. Calculations that assume an activity of 1 over-predict the driving force and place the precipitation temperature too high. Taking an activity of about 0.3 brings the calculated precipitation temperature into agreement with what is found experimentally. For titanium nitride the situation is different: TiN takes very little of the other alloying elements into solution, so its activity can reasonably be treated as close to unity, and its precipitation can be modelled more simply.

For a buyer or a metallurgist, the practical reading is this. Any thermodynamic model of carbide precipitation in H13 that does not account for the multi-component dissolution of the carbide phase will misplace the precipitation temperature, and any process window derived from such a model will be too optimistic about how much of the coarse carbide can be dissolved later. The phases that form close to the solidus cannot be removed by soaking, because the temperatures required would coarsen the austenite grain and damage the surface of the ingot.

Segregation, Cooling Rate and Eutectic Carbide Morphology

The morphology of the eutectic carbide is decided by the cooling rate in the interval between the start and end of the eutectic reaction. At slow cooling the carbide grows as coarse interconnected plates and rods along the interdendritic boundaries, forming a continuous network that is brittle and acts as a crack path. At higher cooling rates the same volume fraction of carbide is distributed as finer, more isolated particles, and the network is broken.

Cooling rate also shifts the precipitation temperature of the eutectic, which is why the measured formation temperature of the carbide varies between ingots and between positions in the same ingot. The centre, the mid-radius and the surface of an ESR ingot are three different experiments in solidification, and a metallographic section taken only from the mid-radius tells an incomplete story.

Position in ESR ingot Relative cooling rate Typical carbide condition Risk in the finished tool
Surface layer Highest Fine, dispersed, thin eutectic Low
Mid-radius Intermediate Mixed network and isolated particles Moderate, orientation dependent
Centre Lowest Coarse continuous eutectic network, largest TiN Highest, cracking and edge chipping

Because the coarse phases are thermally stable, they survive forging and all normal heat treatments. They are retained into the tempered condition, where they act as stress raisers. In a die casting die the consequence is heat checking that initiates at carbide clusters; in a hot forging die it is chipping and gross cracking at the edges and at the corners of the impression.

Measuring Primary Carbide: Rating, Sampling and Acceptance

Acceptance should be written around a rating method, a sampling position and a maximum permitted severity, not around a general statement about cleanliness. Three measurements are usually combined. First, a metallographic rating of the carbide network on a polished and etched section, reported by size class and by whether the network is continuous. Second, a non-metallic inclusion rating to GB/T 10561 or ASTM E45 for the oxide and sulphide content, since large oxides nucleate cracks in the same way as coarse carbide. Third, a grain size determination to GB/T 6394 or ASTM E112 after the specified austenitising treatment, because vanadium in solution controls grain coarsening and its effectiveness depends on how much vanadium was consumed by primary carbide.

The sampling plan matters as much as the method. Sections should be taken from the centre and from at least one off-centre position of a forged bar representing the ingot, oriented so that the carbide stringers are seen in both longitudinal and transverse section. A transverse section is essential: a network that looks broken up on a longitudinal polish can appear continuous end to end across the transverse face.

For hardness verification, use GB/T 230.1 for Rockwell testing on the quenched and tempered specimen, and GB/T 231.1 for Brinell testing on annealed delivery stock. Hardness alone does not reveal coarse carbide, because a die with a coarse network and a die with a fine distribution can both test at 50 HRC; the difference appears in toughness and in thermal fatigue resistance.

ESR Practice Parameters That Refine the Carbide Structure

Because the coarse phases come from the last liquid, every practical lever aims at either reducing segregation or increasing the cooling rate.

Melting rate. A high melting rate deepens the metal pool and enlarges the mushy zone, which lengthens the local solidification time and coarsens the eutectic. Reducing the melting rate flattens the pool, shortens the mushy zone and refines the structure, at the cost of productivity. The economic optimum is the highest rate that still keeps the pool depth within the limit set for the ingot diameter.

Filling ratio and slag pool depth. The ratio of electrode diameter to mold diameter controls the radial heat balance. Too small a filling ratio leaves a wide slag annulus and an unstable pool; too large a ratio reduces the slag thickness and can cause arcing. Slag depth controls the residence time of droplets and therefore the opportunity for oxide absorption, but an excessive depth overheats the pool.

Slag chemistry and oxygen potential. The slag must absorb oxides and sulphur without feeding oxygen back to the metal. A slag with a controlled lime or magnesia addition reduces the oxygen potential and limits reoxidation of titanium, which in turn limits the TiN population. Slag that has picked up too much silica raises the oxygen potential and increases both oxide inclusions and nitride nucleation.

Cooling water and mold design. Higher water velocity on the copper mold raises the extraction rate at the ingot surface and steepens the thermal gradient, which is the single most effective way to break the eutectic network. Mold design that avoids a slack water channel at any height prevents local hot spots where the pool deepens.

Electrode quality. The consumable electrode should be made from clean, low titanium scrap with controlled nitrogen, because titanium and nitrogen entering the process cannot be removed by the slag in useful amounts.

Forging and Heat Treatment Response of Coarse Carbide Stock

Hot working can break a continuous network but cannot dissolve it. Forging practice for H13 stock usually starts from a soaking temperature in the region of 1100-1150 C, with a finishing temperature kept above roughly 900 C to avoid working in the low ductility range, followed by slow cooling. Reduction ratio and forging direction determine how much the network is mechanically dispersed; a single-direction press forging leaves bands of carbide that later appear as property anisotropy.

Annealing for delivery is carried out in the range of about 840-870 C followed by controlled slow cooling to a maximum hardness typically specified around 229 HBW for this grade. Austenitising for service hardness is done around 1000-1040 C, followed by oil or air quenching and at least two tempers in the 520-600 C range, with the second temper mandatory to convert retained austenite. The double temper is not a formality: in a die with coarse primary carbide, retained austenite transforms under service thermal cycling and adds dimensional instability to an already crack-prone structure.

Where the primary carbide is coarse, the practical response is to accept the limitation and design around it: avoid sharp internal corners at positions where carbide clusters are known to lie, specify a generous fillet, and consider a stress-relief temper after any electrical discharge machining of the cavity, since the recast layer over a carbide cluster is a reliable crack initiator.

Common Misconceptions in Procurement

The first misconception is that remelting automatically means clean steel. Electroslag remelting does reduce oxide inclusions and sulphur substantially, but it does not remove the segregation that produces eutectic carbide; in fact a slow-cooled ESR ingot can carry coarser carbide than a small continuously cast billet of the same analysis.

The second is that a hardness specification controls quality. Two deliveries at 50 HRC can differ by a factor of several in unnotched impact energy, and the difference is largely the carbide and inclusion population.

The third is that a higher annealing temperature will dissolve the coarse phase. The eutectic carbide in H13 is stable to temperatures well above any practical annealing range, and an attempt to soak it out merely coarsens the grain and decarburises the surface.

The fourth is that ultrasonic testing replaces metallography. Ultrasonic inspection finds large internal discontinuities; a eutectic network a few tens of micrometres wide is invisible to it, yet it is the feature that governs die life.

FAQ

Q: What is the primary carbide problem in electroslag remelted H13?
Slow solidification of a large ingot enriches the residual liquid in carbon, vanadium, chromium and nitrogen, so titanium nitride and vanadium-rich MC carbide precipitate near the end of freezing and grow into coarse eutectic networks that cannot be dissolved by later heat treatment.

Q: At what stage of solidification does TiN appear in H13?
Published ESR studies place TiN formation at a solid fraction of about 0.82, with vanadium carbide forming very close to the end point of solidification, once the residual liquid is roughly 15-18% of the volume.

Q: Why is carbide activity taken as 0.3 rather than 1 in thermodynamic models?
The precipitated carbide is not pure VC; chromium, molybdenum and iron dissolve in it, which lowers its activity. Assuming unit activity over-predicts the driving force, whereas an activity near 0.3 brings calculated precipitation temperatures into agreement with measured values. TiN behaves differently because it dissolves very little of the other elements.

Q: What is the composition of H13 to GB/T 1299?
4Cr5MoSiV1 is specified at 0.32-0.42% C, 0.80-1.20% Si, 0.20-0.50% Mn, 4.75-5.50% Cr, 1.10-1.75% Mo and 0.80-1.20% V, with phosphorus and sulphur at 0.030% max each.

Q: Which ESR parameter has the strongest effect on carbide refinement?
Cooling rate at the solidification front, driven mainly by mold cooling intensity, together with a controlled melting rate that keeps the metal pool shallow. A high melting rate deepens the pool, lengthens local solidification time and coarsens the eutectic network.

Q: Can forging remove the coarse eutectic network completely?
No. Hot working can break up and disperse a continuous network, and a sufficient reduction ratio improves distribution, but the coarse particles themselves remain. Acceptance therefore has to be based on a metallographic rating at defined sampling positions rather than on reduction ratio alone.

Send Inquiry

whatsapp

phone

Email

Inquiry