Abstract
An important factor to enhance the efficiency and productivity of advanced manufacturing processes is to improve the wear resistance (WR) of tool steels. Cryotreatment in recent years has established itself as a relatively inexpensive additional step in conventional heat treatment schedule that can substantially improve the WR of tool steels. This report highlights the state of understanding of this emerging process with an emphasis on illustrating the underlying mechanisms responsible for the enhancement of WR of tool steels. Cryotreatment almost completely removes retained austenite and induces favourable precipitation of refined secondary carbides with improved distribution. These benefits, if harnessed suitably, are capable of providing significant improvement in both productivity and product quality leading to an efficient economy of manufacturing.
Keywords
Introduction
The manufacturing industry has a continued demand to acquire tool steels which can provide improved service life to achieve higher productivity within techno-economical constraints. As a consequence, considerable efforts have been directed by the engineering community to enhance the useful life of tool steel components specifically by improving their tribological properties. One of the approaches to achieve better wear resistance (WR) of these steels is the employment of cryogenic treatment, popularly referred to as cryotreatment. 1 This methodology has been attempted for a long time, but its improved tuning is rejuvenated only over the last two decades.2–5 Consequently, cryotreatment is finding marked applications in different manufacturing industries such as automobile, aerospace, defense, mining and earth moving equipments. 6 Cryotreatment is a class of sub-zero treatment (SZT); where zero refers to the Celsius temperature scale. 7 The significance of SZT has been realised early by technologists associated with the Swiss watchmakers to the current NASA spacecraft engineers. 8 Scientific dealing with SZT of steels has started as early as 1920s primarily to reduce high level of retained austenite (RA) in the conventionally heat treated (CHT) tool steels. 9 It is a relatively inexpensive, eco-friendly, and one-time permanent treatment that affects the entire cross-section of the component unlike say, coatings. 10
In the manufacturing industry, cryotreatment is specifically aimed to improve a set of properties particularly WR and to achieve improved dimensional stability of tool steel components. This treatment can provide better quality of the products without affecting their yield and thus it remains as one for the emerging needs of several sectors in the present day manufacturing industry. Suitable employment of cryotreatment depends on the understanding of the microstructural evolution and the associated improvement in mechanical properties together with careful analyses of the techno-economic aspects. The underlying theme of this article is to present the current state of understanding of cryotreatment of tool steels. It is pertinent to mention here that cryotreatment has also been applied to various class of materials such as structural steels,8,11 cast irons,12,13 wide range of non-ferrous alloys,14–16 ceramic materials, 17 polymeric materials, 18 metal matrix composites, 19 ceramic matrix composites,20,21 polymer matrix composites, 22 bulk metallic glasses 23 and different nano-tubes 24 /wires 25 /particles 26 for different purposes; these aspects are not dealt in this article.
Process of cryogenic treatment
On the basis of the lowest temperature, SZT can broadly be classified as cryogenic treatment and cold treatment (CT). The temperature range usually associated with cryogenic treatment is from <193 to 77 K, while that related to CT is from <273 to 193 K. 27 In addition to the difference of temperature range, advanced cryogenic treatment, unlike CT, is carried out by using computer controlled cryo-processors with close adherence to the pre-determined cooling and heating rates as well as the preset duration of holding at the lowest temperature. 28 Based on the temperature range, cryogenic treatment has been further classified as shallow cryogenic treatment (SCT) and deep cryogenic treatment (DCT) with an approximate demarcation temperature of 113 K. 29 In summary, the usual temperature range for CT, SCT and DCT are 273–193, 193–113 and 113–77 K, respectively.
The benefits of the cryotreatment can be fruitfully utilised if one carefully selects the associated processing parameters that effectively determine the microstructure and hence, mechanical properties of ferrous materials. The commonly employed sequence of steps in cryotreatment of ferrous materials is shown in Fig. 1a. It is worthy to mention here that cryogenic processing (CP) is neither a complete treatment by itself, nor is a replacement for hardening and tempering treatments. It is only a supplementary step to the conventional hardening and tempering treatments of tool steels.
2
In other words, CP in combination with the hardening and tempering treatments is referred to as cryotreatment. The variables related to a typical CP cycle are: (a) lowest quenching temperature (TLQ), (b) holding time at TLQ, and (c) cooling and heating rates between ambient temperature and the pre-determined cryogenic temperature (Fig. 1b).
a Common sequence of the heat treatment steps in the cryotreatment; b typical time-temperature profile of CP cycle. Processing details of cryotreatment of tool steels
2
The SZT originates from the need to reduce the soft and unstable RA after CHT of tool steels since the characteristic martensite finish (Mf) temperatures of these steels are well below room temperature.30,31 The Mf for most of the commercial tool steels are in the range of 203–153 K,
32
and hence, SCT is sufficient enough to convert almost all RA to martensite.
33
This phenomenon naturally questions the need for further lowering of temperature to that of DCT. The usefulness of DCT lies in the improvement of WR of steels, which is attributed to ‘low-temperature conditioning of martensite’, that, in turn, favours precipitation of finer carbides.2,10,30,34–36 A considerable increase in hardness, WR and service life of tool steels is known to occur by decreasing TLQ down to 77 K (Fig. 2).
36
Effects of lowest quenching temperature (TLQ) and position of sub-zero processing cycle (SZPC) with respect to conventional hardening and tempering treatments on hardness (left axis) and performance of taps (right axis) as measured by numbers of holes drilled for AISI W9 high speed steel, reconstructed following Yan and Li.
36
Here, ‘before tempering’ means SZPC is applied after conventional hardening (from 1503 K) followed by single tempering treatment (at 833 K), whereas ‘after tempering’ infers that SZPC is applied after conventional hardening and triple tempering. Data at TLQ = 303 K means without any SZPC
The holding time at TLQ is one of the most debated parameters for cryotreatment of tool steels. The practice of cryotreatment has been accomplished by different investigators considering a wide range of holding time at DCT (tDCT) which varies from a few minutes to days. Das et al.
37
are the first to perform a systematic investigation related to the influence of tDCT (0–132 h) on the microstructure and mechanical properties of AISI D2 steel. These researchers have demonstrated the presence of an optimum tDCT (≈36 h) for maximising the improvement in WR (Fig. 3). Subsequent theoretical and experimental investigations on different grades of tool steels validate the presence of the optimum tDCT and interestingly, identify the optimum tDCT as around 36 h.38–40 The magnitude of tDCT influences the low-temperature conditioning of martensite which leads to favourable modification in the precipitation of secondary carbides (SCs) during tempering after CP.37,40 The optimum tDCT is the one that results in the desired characteristics of SC particles responsible for improved mechanical properties of tool steels.2,37
Influence of time of holding in DCT (tDCT) on wear resistance of AISI D2 steel specimens tested under different normal loads at constant sliding velocity of 2 m s−1, reproduced from Das et al.
37
The success of cryotreatment is governed by the employed rate of cooling from room temperature to the pre-determined TLQ and heating back from TLQ to the room temperature. Abrupt cooling to low temperature induces considerable thermal stresses as well as stresses arising from austenite to martensite transformation; these stresses lead to micro-cracking. To avoid this problem, usually cooling rates are controlled within the range of 0.3–1.2 K min−1. 2 Rapid heating of predominantly martensitic structure can also lead to cracking. The recommended heating rate (≈0.64 K min−1) is usually lower than that used for cooling. The heating rate should be fast enough to lower the overall time and cost of the process; however, it should be slow enough not to cause cracking. The recommended rates depend on the nature of the steel and the size and shape of the components.
The CP cycle is an add-on process to CHT of tool steels (Fig. 1a), therefore, the details of hardening and tempering are also important together with the sequence of the treatments. For example, Zurecki 41 has shown that the effectiveness of SZT in improving WR depends on the sequence of processing and its associated parameters. Koneshlou et al. 42 have reported that while SZTs increase WR, the degree of improvement varies with TLQ and tempering treatment after CP cycle for H13 steel.
Since, CP ensures higher transformation of austenite to martensite, one can expect more pronounced influence of cryotreatment if hardening conditions lead to higher amount of RA in the conventional hardening of tool steel; i.e. the selection of higher temperature and longer time of austenitisation as well as employment of less severe rate of quenching. 43 Contrary to this view, it has been observed that higher amount of martensite (i.e. less amount of RA) before the commencement of CP cycle is beneficial for the overall performance enhancement of tool steels by cryotreatment. 2 For instance, Amini et al. 44 have shown that increasing quenching severity from austenitising temperature results in higher improvement of WR of AISI D3 steel; because it increases the amount of martensite which will be conditioned at low temperature during CP cycle. It has been contended that the martensite formed prior to CP, not the virgin martensite formed from RA during CP, is responsible for the enhanced precipitation of refined carbides with more uniform distribution. 44 The higher amount of martensite in quenched components of steels, however, greatly increases the probability of their cracking during the course of CP cycle. Therefore, similar hardening parameters as recommended for CHT of a steel is recommended for its cryotreatment for all practical purposes.
Cryogenically hardened tool steels are hard and brittle, and hence, should be subjected to tempering treatment as immediately as practically feasible. Because DCT is known to accelerate the decomposition of martensite and precipitation of SCs during tempering, the selection of lower temperature and/or shorter duration of soaking is recommended for the tempering of cryogenically hardened tool steels. 2 Furthermore, in order to avoid loss of hardness due to over tempering, cryogenically hardened samples should be subjected to single tempering only,33,45,46 in place of double or triple tempering cycles commonly employed for the conventionally hardened tool steels. 32 These modifications of tempering treatment not only reduces the total processing time but also partly balances the cost incurred due to the incorporation of CP cycle in cryotreatment. In earlier practice, cryotreatment were used to be done on the finished product.1,4 But, it is established now that enhancement of mechanical properties like WR is achieved when CP is done just after quenching and before tempering for several grades of steel (Fig. 2). Some investigators have also suggested introducing snap tempering prior to the commencement of CP cycle to reduce the residual stresses so as to diminish the chance of cracking.2,5 Repeated CP cycles have been observed to improve mechanical properties of tool steels; 47 however, the practice should be avoided considering the complexity and the economics of the process.
Influence of cryotreatment on mechanical properties
The magnitude of bulk hardness is commonly used as one of the benchmarks to assess the correctness of heat treatment; therefore, the effects of different SZTs on the bulk hardness of ferrous materials are widely reported. Numerous pieces of evidence suggest that DCT improves bulk hardness only marginally with respect to CHT.1,3,8,33,36–38,41,44,45,48–52,55 Improvement in microhardness by DCT are also reported for different varieties of tool steels;7,10,41 the degree of improvement in microhardness is, however, higher than that of the bulk hardness. Both bulk and microhardness values of steels get considerably affected by TLQ and tDCT. 10 The extent of improvement in hardness values by DCT over CHT increases with increase in austenitising temperature but decreases with increase of tempering temperature. Moreover, DCT either removes secondary hardening peak2,49 or lowers the corresponding temperature,2,10,50 since this treatment accelerates the decomposition of martensite in addition to reduction or even almost complete elimination of RA. 48 The increase in hardness by SZT naturally implies that strength should also be enhanced by this treatment. The reported results by Koneshlou et al., 42 Dong et al., 47 Jaswin and Lal, 52 and Farhani et al., 53 are in line with this expectation except the one by Bensely et al. 54
For AISI D2 steel, Nanesa and Jahazi 56 have reported simultaneous enhancement of strength and ductility along with higher hardness and elastic modulus by DCT with reference to CHT. Improvement of strength properties by DCT has been attributed to the enhanced austenite to martensite transformation and the formation of a higher amount of finer SCs with improved homogeneity. The latter phenomenon also reduces the level of carbon in the martensite when compared with the conventionally treated ones. Therefore, the tempered martensite matrix is relatively soft for the cryotreated tool steels. This along with CP-induced partial dissolution of hard and brittle primary carbides (PCs) is presumed to be responsible for the higher than expected ductility considering the reduction of soft RA content by DCT. 56 Vahdat et al. 51 have also shown that proper selection of parameters associated with cryotreatment results in simultaneous enhancement of hardness and toughness of 45WCrV7 tool steel.
Podgornik et al. 57 have shown that the type and chemistry of tool steels determine the influence of DCT on their mechanical, tribological and load-carrying properties. These researchers have demonstrated that for low carbon cold-work steel, DCT can substantially improve fracture toughness and tribological properties while maintaining a high level of hardness. In contrast, DCT is found to have adverse influence on high-carbon cold-work steel while it has no effect in the case of high-carbon grade of high speed steel for all practical purposes. The observed dependency between the steel composition and the DCT has been attributed to the state of microstructure prior to the commencement of CP cycle, which effectively determines the extent of plastic deformation of primary martensite due to transformation of RA to martensite; here, primary martensite refers to the martensite which is present after CHT. Podgornik et al. 57 have proposed that the positive effect of plastic deformation of the martensite, induced by DCT, diminishes for tool steels with higher amount of undissolved PCs or with higher amount of more stable carbides leading to impairment of properties of tool steels after DCT.
The influence of cryotreatment on fatigue life of various steels has been reported by several groups of researchers. Singh et al. 58 have shown that DCT improves fatigue life of AISI 304L welded joints in the high-cycle regime compared to the similar materials in the untreated condition. The improvement in fatigue life by DCT has been attributed to the formation of martensite during cryotreatment that relieves residual tensile stress and induces compressive stresses around the welded part; as a result, fatigue crack initiation life is improved though the fatigue crack propagation characteristics remain unchanged. 58 Similar results have been reported for AISI 4340 steel, 8 302 stainless steel, 59 and AISI 304 stainless steel in the high-cycle regime (>107 cycles). 60 However, cryotreatment is found to be detrimental for fatigue properties of carburised steels because it causes a reduction in RA content and leads to possible formation of microcracks. 61
There is little consensus on the effects of different categories of SZTs on toughness of steels. He et al. 62 have suggested that incorporation of CT (200 K) for maraging steel in between solutionising (1133 K, 1 h) and aging treatments (773 K, 4 h) can produce excellent combinations of very high tensile strength (2700 MPa), high hardness (62 HRC) along with relatively high level of fracture toughness (30 MPa m1/2), which is tough to achieve even by complex thermo-mechanical treatment. Considerable improvement in apparent fracture toughness by DCT over CHT has been reported for AISI H13 steel 63 and Vanadis 6 steel; 64 however, DCT causes only a marginal reduction in the apparent fracture toughness of PM S390 MC high speed steel. 65 One of the reasons behind such wide scatter in the reported effect of SZT on toughness of tool steels is the inherent variations of processing parameters employed by different investigators.
Perez and Belzunce
66
have investigated mechanical properties of AISI H13 steel subjected to different cryotreatments as well as conventional heat treatments. These investigators have shown that DCT can substantially (24%) improve the fracture toughness of tool steel without affecting other mechanical properties like hardness and strength. It is to be noted that DCT reduces the soft RA content in tool steels, and thus, one should expect only an adverse effect of DCT on fracture toughness. Improvement in fracture toughness of steels by DCT has been attributed to the refinement of SCs and their improved homogeneous distribution in addition to the development of tougher tempered martensite matrix. The latter originates from the consequence of the reduction in the level of carbon in martensite phase by CP coupled with enhanced precipitation of SCs.
66
Das et al.
67
have demonstrated that SZTs, in general, reduce fracture toughness of D2 steel as compared to the CHT ones and that the degree of reduction in fracture toughness varies with the type of SZTs (Fig. 4a). They have explained the observed variation of fracture toughness values of SZT specimens compared to CHT ones by identifying the micromechanisms of fracture; fracture initiates by cracking of PCs as well as by nucleation of microvoids by decohesion of SCs, which, in turn, triggers cleavage fracture.
67
The growth of microvoids is assisted by the presence of RA in the case of CHT specimens (Fig. 4b); while it is assisted by an improved distribution of refined carbide particles in DCT specimens (Fig. 4c).
a Variation of fracture toughness; b fracture surface of CHT specimen; c fracture surface of DCT specimen Variation of chevron notched fracture toughness with lowest quenching temperature (TLQ) for AISI D2 steel specimens, and SEM micrographs of fractured surfaces of CHT and DCT specimens showing microvoids (marked by arrows) formed by decohesion of SC particles. Note the presence of numerous well defined small voids in c, and a few large but well grown voids in b. Reconstructed following Das et al.
67
Improvement in dimensional stability through reduction or elimination of RA is one of the primary reasons to attempt SZT of steels at its inception, and it is established by now that SZT, particularly DCT, enhances the dimensional stability of steel components.1,28,68,69 Recently, cryogenically treated integral diaphragm pressure transducers have been considered for space applications as cryogenic treatment enhances dimensional stability, reduces RA content and decreases the level of compressive residual stresses of a diaphragm made of hardened martensitic (APX4) steel. 68 Transformation of RA also causes significant influence on the magnitude of residual stresses. Surberg et al. 70 have investigated the variation of residual stress with the depth of carburised 21NiCrMo2 steel specimens subjected to CHT, SCT and DCT, and have shown that SCT provides the maximum compressive stress in the case region. Bensely et al. 69 have shown that prior to tempering, the magnitude of compressive residual stresses in the case regions of carburised En 353 steel is in the increasing order of CHT, CT and DCT, but the order gets reversed after tempering.
On the enhancement of WR by cryotreatment
The useful service life of components made of tool steels is determined primarily by their WR. The most prevalent benefit of cryotreatment on these materials is the enhancement of their WR. Barron 71 has made one of the earliest comprehensive studies related to the influence of different types of SZTs on WR of a large variety of ferrous materials. He has shown that in comparison to CHT, WR of several tool steels gets significantly improved by both CT and DCT; however, the extent of improvement by DCT is considerably higher than that by CHT. Similar findings have been reported by numerous researchers such as Amini et al., 38 Oppenkowski et al., 39 Collins and Dormer, 45 Dhokey and Nirbhavne, 46 Podgornik et al., 57 Das et al., 72 Meng et al., 73 Akhbarizadeh et al., 74 Vimal et al. 75 and Thornton et al. 76 Das et al. 72 have demonstrated that SZTs improve WR of AISI D2 steel considerably; the degree of improvement in wear resistance (IWR) varies in the ascending order for CT, SCT and DCT but the extent of IWR reduces with increasing severity of wear test conditions, e.g. increasing applied normal load. It has also been shown that the tDCT determines the extent of IWR of tool steels by DCT (Fig. 3). 77
Considerable IWR by DCT over CHT has been widely reported for various materials like cold-work die steels,39,45,53,64,72–74,77 hot work die steels,42,63,71 high speed steels,36,47,57,78,79 case carburised steels59,61,69,80 and bearing steels.
40
Several investigators36,78,79,81 have also assessed the improvement in service performance of steel components by cryotreatment under simulated laboratory tests. Typical results from the work of Arslan et al.
81
are depicted in Fig. 5. These researchers have evaluated the service life and performance of punches made of AISI D3 steel after cryotreatment by cold-shearing tests on AISI403 stainless steel sheet. This type of punches is widely used in the metal forming process for manufacturing of holes, wherein the tools are usually subjected to severe impact loading and wear during service. It has been reported by Arslan et al.
81
that the service life of punches improves markedly since the observed flank wear can reduce over three times as compared to CHT (Fig. 5a). The common wear phenomenon associated with the degradation of punch life is found to be significantly less in punches treated by DCT when compared with that subjected to CHT after the same number of strokes (Fig. 5b vis-a-vis Fig. 5c). Firouzdor et al.
79
have reported that tool life of M2 HS steel drills when subjected to DCT, increase by 126% in laboratory high speed drilling tests on normalised CK40 carbon steel. Several investigators1,2,5,48,76 have concluded that DCT can yield significant improvement in both productivity and product quality and hence, overall machining economy, offsetting the cost of PC cycle.
a Volumetric flank wear; b micrograph of CHT specimens; c micrograph of DCT specimens. Evidences related to the improvement of life and performance of punches (diameter 5 mm) made of AISI D3 steel by cryotreatment (DCT, with holding time of 36 h at 128 K and tempered for 2 h at 423 K) with reference to CHT during punching of AISI 403 stainless steel sheet following Arslan et al.
81
The markings in SEM photographs (bottom set) corresponding to different identified wear mechanisms: A – fretting fatigue; B – cavitation abrasive wear; C – abrasive (grinding) wear; D – shear or plastering wear; E – adhesive wear and F – diffusion wear
While it has been established that DCT improves WR of tool steels, the achievable degree of improvement in WR by DCT is still debated. This is because the reported IWR by DCT over CHT is inconsistent;82,83 it varies widely from a few per cent to several hundred per cent for the same material even under laboratory test conditions.41,55 This uncertainty was limiting the wide spread commercial applications of cryotreatment, but the debate on the degree of benefit of CP of tool steels appears to have been resolved by Das et al.10,84,85
Das et al.84,85 have performed a systematic and in-depth investigation on dry sliding wear behaviour of AISI D2 steel specimens subjected to CHT and DCT. Apart from the measurement of wear rates over a wide range of normal loads (FN), these researchers have identified operative modes (mild/severe) and mechanisms (oxidative/metallic) via detailed characterisations of the generated wear debris, surfaces and sub-surfaces of the worn specimens. It has been observed that both types of specimens exhibit an abrupt increase in wear rate at a particular FN (Fig. 6a) which is associated with the change in the operative mode and mechanism of wear from mild oxidative to severe-metallic. This phenomenon is known as mild-to-severe wear transition, and the particular FN value is commonly referred to as T1 transition load.
86
Das et al.
85
have demonstrated for the first time that cryotreatment considerably increases the T1 transition load of tool steels (Fig. 6a). Mild-to-severe wear transition is usually associated with the dynamic changes occurring at the contact surfaces and sub-surfaces during the course of wear, e.g. the generation of the detrimental white etching layer in tool steels is due to the combined actions of frictional shear stress and frictional heat as determined by the imposed wear test conditions.
87
The favourable microstructural modifications imparted by DCT offer a Wear rate; b degree of IWR by DCT over CHT Variations of the wear rate and the degree of IWR (β) by DCT with respect to CHT for specimens of AISI D2 steel as functions of applied normal load during sliding wear. Reproduced from the work of Das et al.
84
See text for further detailss higher resistance to plastic deformation and thermal softening during wear and, as a consequence, wear transition occurs at more severe wear condition like higher load and/or sliding velocity for cryotreated tool steel.
In order to explain the reported uncertainty related to IWR by DCT over CHT, Das et al.
84
have first estimated the WR using the measured wear rate, applied FN and Vickers hardness of the test specimen employing the well known Archard's law of wear.
88
The ratio (β) of the calculated values of WR of CHT (WRCHT) and that of DCT (WRDCT) specimens has been considered as an index of the degree of IWR by DCT over CHT following Barron.
71
Das et al.84,85 have proposed a two-dimensional map depicting the variation of the magnitude of β as a function of FN (Fig. 6b) which helps to understand the wide variations of IWR by DCT over CHT. Figure 6b reveals that the magnitude of β varies widely with FN, but are confined to three distinct zones. The values of β are found to be only a few times in the zones-I and III where the mode and mechanism of wear operative in CHT and DCT specimens are identical; i.e. mild oxidative in the zone-I (at lower FN) and severe-metallic in the zone-III (at higher FN). The values of β are over an order of magnitude in the zone-II (at the intermediate range of FN), where the operative modes and mechanisms are different for CHT and DCT specimens, i.e. severe-metallic for the former specimen and mild oxidative for the latter specimens (see Fig. 7). Das et al.82,84,85 have explained that the reported variation in the degree of β is due to either similar or dissimilar modes and mechanisms of wear being operative in CHT and DCT specimens, controlled by the imposed wear test conditions.
a and b Worn surfaces; c and d generated wear debris; e and f sub-surfaces of worn specimens. Evidences of difference in the operative modes (mild/severe) and mechanisms (oxidative/metallic) of wear between CHT and DCT specimens, i.e. severe-metallic for CHT and mild oxidative for DCT samples, under identical wear condition (load: 49 N and sliding velocity: 2 m s−1). Reproduced from Das et al.
85
aWith respect to conventional heat treatment.
On the microstructural modification by cryotreatment
Scientific investigations by now have established that cryotreatment can substantially improve the desirable properties and performance of tool steels. The improvements in the properties by cryotreatment have been attributed to one or combinations of the following microstructural alterations: (i) more complete transformation of austenite to martensite,9,33,68–71,79,90 (ii) alteration in the lattice parameters of the martensite like reduction in its tetragonality,34,35,91 (ii) preferential precipitation of transition η-carbides, 73 (iii) precipitation of nano-sized carbide particles 92 and/or formation of nano-sized martensite particles,58,60 and (iv) alterations in the decomposition of martensite leading to enhanced precipitation of refined SCs with more homogeneous distribution.47,53,74,79,81,93–96
Meng et al. 73 have inferred that DCT promotes preferential precipitation of fine η-carbides (also referred to as ε′(ε), Fe2C, orthorhombic 97 ) instead of ε-carbide (Fe2.4C, hexagonal closed packed) which is responsible for simultaneous improvement in hardness and toughness, and, in turn, enhances WR of D2 steel. Several later investigations38,45,70,74,76,94 have not been able to detect the presence of η-carbide in cryotreated tool steels; this is expected because η is a transition carbide and hence, either dissolves or gets converted into stable θ (Fe3C)-carbide or alloy carbides. 43 Proposition related to the formation of nano-sized carbides and/or martensite particles either lacks satisfactory evidence 92 or have limited applicability like only for austenitic stainless steels.58,60 It is well established that CP cycle reduces the tetragonality of martensite34,35,91 and, in turn, raises the toughness of the tempered martensite matrix. But this modification alone cannot account the considerable improvement in mechanical properties like marked enhancement of WR recorded for tool steels. It is established now beyond doubt that DCT does not alter the nature of carbides for a particular material.42,64,94 Several systematic investigations53,56,61,63,64,66,69,74,81,93,98,99 over the last few years help to conclude that the enhancement of properties of tool steels by cryotreatment is due to favourable modification of the characteristics of SCs in addition to complete transformation of austenite to martensite. These aspects and the related micromechanisms are discussed in the subsequent sub-sections.
Role of cryotreatment on transformation and stability of RA
The martensite finish (Mf) temperature of ferrous materials especially those with high-carbon and alloying elements lies well below the room temperature; therefore, conventional hardening treatment fails to transform considerable amount of austenite in to martensite often leading to unacceptable level of RA in the as-quenched structure of these materials. 31 The RA is soft, and it adversely affects the desirable properties like hardness and WR. In addition, RA is unstable and can transform into martensite during service of components made of tool steels, which is undesirable since the freshly formed martensite being untempered is brittle in nature. Transformation of austenite to martensite is further associated with approximately 4% volume expansion, which leads to dimensional changes, distortion of the components and even causes failure in extreme cases.31,32,43,92 Therefore, SZTs have been recognised as early as 1920s for reducing the amount of RA in ferrous materials.2,9,27
The improvement in mechanical properties of ferrous materials due to SZTs are traditionally attributed to either partly36,45,47,53,66,74,78,81 or solely9,33,49,71 to the reduction of RA content. The latter proposition, however, cannot explain the recorded variation in properties processed by different SZTs. One should not expect any appreciable difference in the level of RA content of specimens treated by SCT and DCT (see Fig. 8) since the TLQ of SCT itself is below the characteristic Mf temperatures for most commercial steels.31,43
Typical examples related to the influence of SZTs on microstructural modifications in AISI D2 steel. Reproduced from the work of Das et al.
67
CHT: conventional heat treatment, i.e. without any sub-zero processing cycle; CT: cold treatment; SCT: sallow cryogenic treatment; DCT: deep cryogenic treatment; SCs: secondary carbides
The extent of reduction of RA content in ferrous materials by SZTs is still debated. A group of investigators36,47,50,74,78,84 have claimed that DCT almost completely removes RA (to <2 vol.-%, the detection limit of XRD technique), while several others34,42,53,54,61,65,68,73 have reported that DCT indeed reduces RA content considerably but it fails to remove it completely. This controversy initiates from the difference in the selected TLQ with regard to the Mf temperature of a selected steel apart from the other processing parameters of CP cycle. Further, thermal stability of RA can reduce its subsequent transformation which is dependent on the sequence of the treatments and the choice of the associated process parameters; notably, time and temperature of austenitisation, rate of quenching and time of holding of conventionally hardened specimens at ambient temperature before the commencement of CP cycle.2,10,74
Another controversial issue related to the austenite to martensite transformation during sub-zero processing is the nature of transformation. It is widely believed that transformation of austenite to martensite in Fe–C system is athermal in nature. 43 However, Gavriljuk et al. 100 have observed isothermal transformation of austenite to martensite for high-carbon high-alloy tool steels. It has been reported that isothermal transformation of austenite occurs at temperatures higher than 77 K but lower than 173 K, and the maximum isothermal kinetics is observed at around 123 K due to compromise between the driving force for transformation, being proportional to the decrease in temperature, and the thermal activation needed for the isothermal transformation. For 52100 bearing steel, Preciado and Pellizzari 101 have observed that while the majority of RA transforms to martensite immediately after reaching cryogenic temperature (93 K), prolonged holding causes further transformation, which is indicative of thermally activated transformation of austenite to martensite. The RA in cryotreated tool steels exhibits higher thermal stability; because it possesses higher compressive residual stresses, which are generated due to enhanced transformation of austenite to martensite in CP. This is supported by the fact that activation energy related to the decomposition of austenite is found to be higher in DCT samples over CHT ones. 101
Villa et al.
102
have studied the evolution of compressive strain associated with transformation of austenite during CP of tool steels by means of in situ and ex situ (synchrotron) X-ray diffraction techniques. It has been inferred by them that austenite to martensite transformation continues during all stages of CP cycle, i.e. cooling, isothermal holding and reheating stages. The development of martensite during isothermal holding and reheating stages is interpreted as thermally activated growth of thin martensite plates, generated athermally during cooling stage. Continued transformation of austenite to martensite builds up compressive strain in the remaining austenite when the transformation is above a certain critical temperature, which is around 140 K for 52100 steel. Below the critical temperature, there is no build-up of compressive strain in austenite, since the martensite itself accommodates the transformation strain by internal twinning. The presence of twins in the martensite has been reported by several investigators for the cryogenically hardened steels,94,103,104 (see Fig. 9). Villa et al.
102
have also suggested that the small amount of austenite which is retained even after CP is mechanically stabilised; because small regions of RA are encapsulated within martensite (Fig. 10). CP is also found to accelerate the decomposition of RA to ferrite plus cementite during tempering at elevated temperatures; the decomposition of austenite starts in cryogenically hardened 52100 steel at about 30 K lower than that observed for conventionally hardened specimens. Based on results obtained from highly sensitive dilatometry tests and detailed microstructural characterisations of cryogenically treated AISI D2 steel, Nanesa and Jahazi
104
have proposed an alternative phase transformation path instead of the traditionally expected martensitic transformation from austenite. It has been reported that the austenite transformation starts with carbide precipitation (1 μm2) and it continues with the formations of allotriomorphic ferrite, Widmanstätten ferrite, acicular ferrite and finally bainite. These observations necessitate re-examination of the role of SZTs on phase transformation as well as on the tempering response of cryotreated tool steels.
a CHT; b DCT. Bright-field TEM micrographs illustrating the difference in microstructures of Vanadis 6 steel after conventionally and cryogenically (77 K, 4 h) hardened state. Reproduced from the work of Jurci et al.
103
Backscatter electron micrograph of electrolytically etched AISI 52100 steel subjected to cryotreatment (austenitised at 1353 K for 60 s, oil quenched to 413 K and holding for 20 s, air cooled to room temperature, heated to 453 K at 1.5 K min−1, cooled to 77 K at 5 K min−1 and finally heated to 300 K at 1.5 K min−1) following Villa et al.
102
RA regions appear darker and embedded in between the martensite plates; new martensite is fine and appears lighter than the martensite developed prior to CP
On the modifications of SCs by cryotreatment
Focused research in recent years has helped to establish that SZT, in general, and DCT, in particular, not only enhances the transformation of RA to martensite but also imparts favourable modifications in the characteristics of SCs.
105
The latter phenomenon is a manifestation of what is popularly termed as ‘low-temperature conditioning of martensite’.2,10,45,93,99,105,106 On the basis of quantitative results on differently heat treated AISI D2 steel, Das and his co-workers90,108 have demonstrated that, in addition to the alteration of RA content, SZTs refine the size of the SCs, increase their amount and population density, and lead to more uniform distribution of SCs in the microstructures. These noteworthy insights have been realised by Das et al.55,90,105,108 via categorisation of carbide particles as PCs (size >5 µm) and SCs (size ≤5 µm) followed by sub-classification of the latter as large SCs (size of LSCs; 1 µm < size ≤ 5 µm) and small SCs (size of SSCs; 0.1 µm ≤ size ≤ 1 µm). These favourable modifications of the SCs increase with decreasing TLQ
108
and gets enhanced with increasing tDCT at 77 K up to a critical duration (Fig. 11).77,90 Typical representative SEM (Fig. 12) and TEM (Fig. 13) micrographs of AISI D2 specimens subjected to CHT and DCT treatments illustrates that DCT indeed significantly modifies the characteristics of SCs in tool steels. The favourable alterations of the carbide particles are one of the major microstructural changes by cryotreatment. This has been re-affirmed later on by several investigators in various grades of tool steels.35,36,53,66,79,81,91 For example, Farhani et al.
53
have reported for 1.2542 tool steel that, in comparison to CHT, DCT significantly increases the population density of SSCs. Firouzdor et al.
79
have shown that the amount of fine carbides in DCT specimens is 10 vol.-% against only 6 vol.-% in CHT specimens. Based on synchrotron diffraction studies on AISI H13 tool steel, Xu et al.
35
have shown that while both treatments result in the same V-rich cubic M8C7 carbide, its concentration increases by DCT as compared to that by CHT.
Variations of mean diameter of small SCs (SSCs) and large SCs (LSCs) with time of holding in DCT (tDCT) at 77 K for AISI D2 steel following Das et al.
90
a CHT; b DCT. Representative SEM micrographs of AISI D2 steel illustrating difference in the amount, size and distribution of small SCs (SSCs) and large SCs (LSCs) in CHT and DCT specimens. Region of RA is present only in CHT sample. Reproduced from the work of Das et al.
108
a CHT; b DCT. TEM micrographs exhibit difference of size and distribution of ultrafine SCs in CHT and DCT specimens of AISI D2 steel. Reproduced from the work of Amini et al.
106
Dhokey et al. 107 have inferred that the cryotreatment influences nucleation and growth of carbides. For M35 tool steel, the nucleation of carbides dominates for the first 16 h of isothermal holding at 88 K resulting in higher carbide density and lower residual stresses; whereas, the growth of carbides dominate beyond this time, leading to higher residual stresses. Dhokey et al. 107 have also suggested that relatively large stress relaxation in the matrix during heating stage of CP cycle causes nucleation of fine incoherent carbides resulting in higher carbide density.
Preciado and Pellizzari 101 have examined the thermal decomposition of martensite in 52100 steel subjected to CP at 93 K for short (5 min) and long (24 h) durations with reference to CHT ones employing differential scanning calorimetry and dilatometry. These investigators have shown that DCT enhances enrichment of carbon atoms and formation of atomic clusters in the pre-precipitation stage of tempering of martensite. The activation energy, associated with the above phenomena, is found to be lower for the cryogenically hardened steel as compared to the conventionally hardened ones indicating faster redistribution of carbon in the former. The extent of reduction of activation energy increases with increasing time of holding at 93 K. 101 This phenomenon has been attributed to the conditioning of martensite at cryogenic temperature. Conditioning of martensite infers generation of higher density of crystal defects like dislocation and twins which act as preferential sites for the segregation and clustering of mainly carbon atoms.
Gavriljuk et al. 91 have reported that no fine carbides are precipitated during prolonged isothermal holding at cryogenic temperatures as one expects immobility of carbon atoms at such low temperatures. During tempering in between 373 and 473 K, precipitation of transition η-carbide has been detected in conventionally as well as cryogenically treated steel specimens when CP is carried out at 73 K. However, no such transition carbide is observed when CP is done at 123 K; this has been explained by the intensive isothermal martensitic transformation and its associated plastic deformation. 91 Furthermore, precipitation of Fe3C and its coarsening, and absence of carbides of alloying elements are the general features of cryogenically hardened tool steels during tempering up to 773 K; in contrast, precipitation of Fe3C starts at 573 K followed by its dissolution, and subsequent precipitation of carbides of alloying elements at 773 K for conventionally hardened ones.
It is considered that the enhanced precipitation of refined carbide particles and their improved homogeneous distribution in tool steels by cryotreatment are the consequence of CP-induced generation of atomic clusters that grow into or act as heterogeneous nuclei for the formation of carbide particles on the subsequent stage of warming or during tempering. About the atomistic mechanisms behind the formation of these atomic clusters, there exist two contradictory hypotheses.
A group of researchers94,96,103,106,109 has suggested that high density of crystal defects such as dislocations and twins get generated in martensite when sub-zero cooling is done, which results in high internal stresses during the transformation of austenite to martensite as well as from the differential thermal contractions of the various phases. This phenomenon increases the lattice distortion and thermodynamic instability of martensite which results from the segregation of carbon atoms to nearby defect forming clusters. This hypothesis considers defect-induced enhanced diffusion of carbon atoms to nearby crystal imperfections like dislocations and twins especially during the warming up stage. It corroborates well with the widely recorded facts that the extent of modification of carbide increases with decreasing TLQ and increasing tDCT.10,108 This proposition is also well supported by the findings of Li et al. 109 based on internal friction characteristics of differently cryotreated cold-work die steel as well as by the observations of Akhbarizadeh and Javadpou 110 on the role of as-quenched vacancies during the precipitation of carbides by DCT by means of controlling the electric current flow during CP cycle in D2 steel.
Another group of investigators91,100,111 have suggested an alternative mechanism for the formation of atomic clusters due to CP. It is based on the plastic deformation of martensite which is formed during CP cycle, specifically, by isothermal transformation of austenite. It presumes that the martensite formed at cryogenic temperature is sufficiently soft due to immobility of carbon atoms at such low temperatures, which inhibit the clustering of C atoms or the process of auto-tempering of the newly formed martensite. 111 They have argued that an important consequence of plastic deformation of martensite is in its capturing the essentially immobile carbon atoms by moving dislocations leading to the formation of atomic clusters. The unusually low tetragonality of the martensite recorded in the cryotreated tool steels has been forwarded as the evidence for this hypothesis.35,101,102 Further investigations are warranted to resolve the atomistic mechanism related to the modification of SCs by DCT.
Summary and concluding remarks
Cryogenic treatment, a supplementary step to the conventional heat treatment schedule, is relatively an inexpensive and eco-friendly treatment to achieve improved service performance of tool steels with added advantage of being one-time permanent treatment, unlike coating.
Cryotreatment is most effective if CP cycle is incorporated after conventional hardening and before tempering treatment. However, stress relieving treatment before the commencement of CP should be considered for intricate shape of tool steels with higher cracking sensitivity to avoid breakage.
The recommended lowest temperature in CP cycle should be close to 77 K. The rate of cooling/heating from/to ambient temperature to the pre-selected cryogenic temperature should be as fast as possible but slow enough to avoid cracking. The time of holding at the lowest temperature should be optimised; the optimum time appears to be 36 h for several grades of tool steels.
Cryogenically hardened tool steels should be subjected to single tempering treatment instead of commonly employed double/triple tempering treatments. The temperature of tempering should be about 50 K lower than that recommended for a particular steel in case of conventionally hardening.
The major benefit of cryotreatment is significant IWR of tool steels. The extent of IWR is, however, governed by the operative modes (mild/severe) and mechanisms (oxidative/metallic) of wear as dictated by the service conditions such as applied load, relative velocity and state of lubrication, etc. If the modes and mechanisms of wear for conventionally and cryogenically treated steels are similar, which can be either mild oxidative or severe-metallic, the degree of IWR by cryotreatment is only a few times. Whereas, the degree of IWR is over an order of magnitude when the modes and mechanisms are dissimilar. The real benefit of cryotreatment of tool steels gets highlighted when these are used under relatively harsh service conditions.
For tool steels, cryotreatment markedly enhances dimensional stability and induces compressive residual stresses. It increases hardness, tensile and bend strength but the magnitude of improvement varies from marginal to moderate. Moreover, this processing step does not have any noticeable adverse effects on other crucial mechanical properties like fatigue, impact or fracture toughness.
The science behind the improvement of properties of tool steels by cryogenic treatment is now fairly well established. Cryotreatment almost eliminates the RA. This treatment does not alter the nature of carbides, but it significantly influences the precipitation behaviour of SCs in steels leading to increase in their amount and population density associated with more uniform distribution. The extents of these favourable microstructural modifications are functions of time and temperature of CP cycle apart from the chemistry and the prior thermal history of materials.
Two different hypotheses exist regarding the characteristic alterations of carbide particles during cryogenic treatment. One is based on defect-induced enhanced diffusion of carbon atoms to nearby crystal imperfections while the other considers plastic deformation of virgin martensite during CP as the cause for the formation of atomic clusters which act as nucleating sites for precipitation of carbide during tempering of cryogenically hardened steels.
The mistrust surrounding the benefit of cryotreatment, specifically for tool steels, has disappeared by the outcomes of several useful pieces of research over the last two decades. Research efforts should now be directed to optimise processing cycle keeping end use in mind, in one hand, and to understand the underlying operative micromechanisms, on the other hand, so as to extract the real potential of this novel processing step.
Footnotes
Acknowledgements
The assistance received from the Centre of Excellence on Microstructurally Designed Advanced Materials Development, TEQIP-II, IIEST Shibpur to carry out a part of this work is gratefully acknowledged.