Introduction

AISI 440 C is a high-carbon martensitic stainless steel with superior wear and abrasion resistance among all the grades of stainless steel. 440C stainless steels are extensively used in aerospace for ball bearings and hydraulic control valve assemblies [1,2,3]. Corrosion resistance is an important consideration for hydraulic valve applications. Even though high carbon content aids the functional needs of the steel, corrosion resistance reduction takes place by the formation of primary and secondary carbides of chromium [4, 6,7,8]. The chromium content (18wt.%) in AISI 440C steel is equivalent to corrosion resistance of austenitic stainless steel, but its corrosion resistance becomes inferior to austenitic grade due to most of the chromium being shared by carbon; there is no extra chromium to heal the damages in the passivated chromium oxide layer. A minimum of 11 wt.% Cr is required to form the chromium oxide film uniformly; chromium content less than 11wt.% can cause localized damage by forming pits [4, 5]. Pits are formed near M7C3 primary carbides due to less chromium in the matrix [3]. The presence of more elemental chromium in the martensite matrix will enhance the corrosion resistance by forming the passive chromium oxide layer [9, 10].

Chlorine-rich atmosphere damages the passive layer and subsequently leads to the formation of pits. Hydrolysis at the pit location goes into autocatalytic mode and pits grow further and further until they perforates [4, 11]. Pit size, shape, and depth depend on several factors. Deep narrow pits act as severe stress concentration in loading. In aggressive corrosive environments, 440C steel has very poor corrosion resistance; in such cases, pitting and corrosion become more dynamic. In several conditions, pits act as initiators for stress corrosion cracking (SCC) and crevice corrosions. Austenitizing temperature plays a role in the control of primary carbide content [12]. Austenite treatment at higher temperatures leads to the dissolution of more primary carbides; as a result, matrix is enriched with elemental carbon. During hardening it will bring down the Ms temperature. As a result, martensite transformation will be incomplete and some fraction of austenite is retained. Austenitizing at lower temperatures decreases the corrosion resistance by retention of more primary carbides; as a result, martensitic matrix depleted with free form of chromium [5, 13]. Considering both these aspects hardening temperature is selected.

Hydraulic amplifier valve (HAV) spools are used in flow control devices of launch vehicle actuator systems. In the current case, failure was observed near the undercut region of the HAV spool during the vibration test (Fig. 1). It failed during the vibration check for use in the flight. A detailed investigation has been carried out to understand the root cause of the failure.

Fig. 1
figure 1

Photograph showing the failed HAV spool with failure location with arrow

Full size image

Experimental

HAV spools are realized through a vacuum arc melting route. Visual and stereomicroscopic observations were carried out on the failed HAV spool. Fractographic analysis was carried out on the fracture surface under Carl Zeiss EV 50 scanning electron microscope (SEM) and chemical composition was analyzed by OXFORDX-MaxN energy dispersed X-ray spectroscopy (EDS). One of the fracture portions was sectioned vertically with the slow-speed diamond cutting machine for optical metallographic analysis. Emery polishing was done using different grades of emery papers. Final polishing was done using alumina followed by diamond paste. A metallographically prepared sample was observed under an optical microscope. Etching was carried out using 1% Nital for 10 s by swabbing. Etched specimens were observed under Olympus GX 71 optical microscope. Microhardness measurements were carried out by using a Wilson hardness machine with a 300gf load. Further, electron backscattered diffraction analysis (EBSD) was also carried out using a field emission scanning electron microscope (FESEM) of Carl Zeiss Gemini 560 with a step size of 0.5 µm and voltage of 20 kV. Buehler Vibromet polishing machine was used with 0.05 µm size colloidal silica solution in the final polishing for 24 h to get the clear Kikuchi patterns. Post-processing analysis was done by using Aztec Crystal software.

Results and Discussion

Visual, Digital and Stereomicroscopic Observations

Visual and stereomicroscopic observations showed fracture occurred at the undercut region of the M5 thread portion as shown in Fig. 2. Both spool side and screw side fracture surfaces were having corrosion products as shown in Fig. 3a and b. The undercut region of the fracture surface revealed rough machining marks (Fig. 3c). Brown color corrosion patch was observed on the spool surface portion, which is slightly away from the fracture edge (Fig. 3d).

Fig. 2
figure 2

Photographs showing fracture surfaces of HAV spool both side and corrosion patch present on the surface

Full size image
Fig. 3
figure 3

Photographs showing fracture surfaces of HAV spool both sides and corrosion patch present on the surface

Full size image

Scanning electron microscopy (SEM) and Energy Dispersed X-ray Spectroscopy (EDS) analysis

SEM observations on the HAV spool fracture surface revealed corrosion products present near to edge locations (undercut region with rough machine marks). Corrosion pit of approx. 612 µm depth was observed in the fracture location (Fig. 4a, b and c). EDS elemental analysis confirmed the presence of oxygen and chlorine in the pit area (Fig. 5). The fracture surface away from the pit region has dimple features with primary carbide fragmentation in a brittle manner along with spherical secondary carbides (Fig. 6). Rough machine marks and small pits were observed at the undercut region of the fracture edge side (Fig. 7). A corrosion patch was observed on the HAV spool body away from the fracture surface (Fig. 8). SEM&EDS elemental analysis revealed the pit morphology and chemistry. Oxygen and chlorine presence were confirmed at the pit location in the corrosion patch. Chlorine concentration at the pit location is high as shown in Fig. 9.

Fig. 4
figure 4

SEM Fractographs showing the Pit region and away

Full size image
Fig. 5
figure 5

EDS analysis at pit location of the fracture region

Full size image
Fig. 6
figure 6

SEM Fractographs showing (a, b) Fracture surface and (c, d) Polished surface near to fracture

Full size image
Fig. 7
figure 7

SEM Fractographs showing pits and rough machining marks

Full size image
Fig. 8
figure 8

Pit location away from the fracture edge with corrosion products

Full size image
Fig. 9
figure 9

EDS analysis on the Pit surface and away for pit

Full size image

SEM micrographs revealed primary and secondary carbides with different sizes and geometry (Fig. 10 a, b, c and d). Primary carbides are in irregular shapes with different sizes-based Cr/Fe ratio. Secondary carbides are in spherical morphology with different sizes, formed during the tempering stage. EDS area map analysis confirmed the chromium-rich phases with high chromium concentration (Fig. 11). Line scan analysis confirmed the Cr variation at the martensitic matrix and carbides (Fig. 12).

Fig. 10
figure 10

SEM micrographs showing the primary and secondary carbides

Full size image
Fig. 11
figure 11

EDS Elemental analysis on the chromium-rich phases and martensitic matrix

Full size image
Fig. 12
figure 12

EDS line scan analysis showing Cr distribution

Full size image

Based on the SEM observations, it is clear that failure has occurred due to a deep pit at the undercut region. During vibration test, this pit acted as a notch and reduced the load-bearing area; hence, fracture occurred at the undercut region. It is well known that primary carbides form during the solidification of the liquid metal. Some of these carbides will be exposed to the surface during the machining stage. The corrosion pits preferentially initiate around the bigger primary carbides due to the potential difference between the carbide and matrix. Aggressively chlorinated atmosphere further damages the chromium oxide layer and encourages the pit initiation and growth. Mechanisms of pitting are not well understood yet but the most believed mechanism is that chloride ions adsorb into the oxide layer at high energy locations (scratch, inclusion, carbides) and diffuse to metal/oxide interface. Since enough chromium is not there in the matrix, iron will start sacrificing and form porous brittle oxides. Chlorine reacts with moisture around it and forms an acid environment in the pit location where further increases the localized attack until it perforates or fails.

Pitting corrosion of martensitic stainless steels is a complex phenomenon. Primary carbides (M7C3) coming through solidification are more noble than surrounding martensite matrix due to their higher Cr content and lower Fe content. It indicates they have higher corrosion resistance than matrix. The exact potential can vary depending on specific carbide type and its associated composition and environment. Primary carbides like M7C3 (Orthorhombic) and secondary carbides like M23C6 (FCC) with different Cr/Fe ratios and crystal structures have slight variations in their corrosion potential.

Factors affecting pitting potential

  1. 1.

    Carbide Type and Its Composition

    Among primary carbides (M7C3), secondary carbides (M23C6) and martensitic matrix, primary carbides will have higher Cr/Fe ratio than the rest two. Martensitic matrix near to primary carbides will have very low Cr/Fe ratio with respect to primary carbide. This will generate a potential difference leading to the initiation of corrosion pits.

  2. 2.

    Heat treatment and Microstructure

    Improper austenite treatment and improper tempering can make martensitic matrix with less Cr/Fe. Austenitizing at higher temperature dissolves most of the primary carbides, and matrix enrich with Cr; hence, pitting resistance improves. Tempering at higher temperature also can deplete the chromium content from the matrix through secondary carbide formation [5].

  3. 3.

    Electrolyte Composition and pH

    The specific environment where the stainless steel is used can significantly impact the pitting potential of both carbides and matrix. Higher concentration and high pH can lead to immediate pitting [4].

Microscopy and Microhardness

Optical microscopic observations on the polished cross section of M5 threaded region side revealed microstructure consists of plate martensite, primary and secondary carbides. Most of the primary carbides align perpendicular to the fracture edge. Some of the carbides get fragmented during fracture. No secondary cracks near the fracture edge and corrosion were observed at cross section change location of the HAV spool as shown in Fig. 13a and b. Primary carbides opened to machined surface and pitting corrosion was initiated near to primary carbides (Fig 13c). During the formation of primary carbides, chromium from the matrix is taken, and subsequently, martensite matrix is leaned with elemental chromium. By any chance if the passivated chromium oxide layer gets damaged, pits will nucleate around the primary carbide since there is not enough chromium to heal the damages. Primary carbides opening to the surface during machining will create problem as the carbides are brittle compounds (Fig. 13c). However, no damage was found in the threads, which were made by machining process (Fig. 13d). Microhardness measurements are in the range of 595-610 VHN.

Fig. 13
figure 13

Optical micrographs of the fracture surface of HAV spool M5 threaded side

Full size image

Pitting corrosion is an extremely localized form of corrosion which ended up with holes in the alloy. These pit holes may be small or big and deep or shallow; it depends on the environment. Deeper pits can create more damage with less weight loss. It is difficult to identify the pitting corrosion at starting stage since pits diameter is in micro-level. It will take time to become visible to the naked eye. Pitting is more dangerous under dynamic loading conditions than static conditions.

EBSD analysis was performed on the specimen away from fracture. BC(Band Contrast) + Ph(Phases) maps have shown the presence of M23C6, M7C3 carbides and small fraction of austenite (Fig. 14b). Primary carbides are bigger in size than secondary carbides. BC + IPF + GB map shows plate martensite oriented randomly (Fig. 14a).

Fig. 14
figure 14

EBSD maps showing (a) BC + IPF + GB (b) BC + Ph

Full size image

Conclusions and Preventive Actions

Detailed analysis was carried out on the hydraulic amplifier valve spool to understand the root cause of the failure. The failure was observed at the undercut region of the M5 thread side. Fracture surface revealed a pit of approximately 612 µm. Elemental analysis using EDS confirmed the presence of oxygen and chlorine in the pit. Microstructure and microhardness are typical of AISI 440C steel in hardened and tempered condition. The pits were attributed to primary carbides exposed on the machined surface. Ultimately, the failure of the HAV spool was attributed to the corrosion-assisted overload. To avoid future failures related to corrosion in aggressive environments following actions are suggested.

  1. (a)

    Regular inspection at inaccessible corners and undercut regions.

  2. (b)

    Corrosion protection using lubrication or keeping in oil,

  3. (c)

    Alternatives of 440C like CR-30 and 15-5PH can be used.