Wear 261 (2006) 674685

Performance of cryogenically treated HSS tools

Fl avio J. da Silva a , Sin esio D. Franco b, , Alisson R. Machado c , Emmanuel O. Ezugwu d , Ant onio M. Souza Jr. e
a

Federal University of Esp rito Santo, Technological Center, Department of Mechanical Engineering, 29.060-970 Vit oria, ES, Brazil b Federal University of Uberl andia, Faculty of Mechanical Engineering, Tribology and Materials Laboratory, LTM, 38.400-089 Uberl andia, MG, Brazil c Federal University of Uberl andia, Faculty of Mechanical Engineering, Machining Teaching and Research Laboratory, LEPU, 38.400-089 Uberl andia, MG, Brazil d London South Bank University, Machining Research Laboratory, Department of Engineering Systems, London, UK e Fiat-GM Powertrain, F.A. Powertrain Ltd., Engineering Manufacturing, Betim, MG, Brazil Received 13 March 2005; received in revised form 23 January 2006; accepted 24 January 2006 Available online 10 March 2006

Abstract Studies on cryogenically treated high speed steel tools show microstructural changes in the material that can inuence tool lives and productivity signicantly. Results in the literature show tool life improvements from 92% to 817% when using the cryogenically treated HSS tools in the industry. However, the real mechanisms which guarantee better tool performance are still dubious. This implies in the need of further investigation in order to control the technique more scientically. This work aims to verify the effect of cryogenic treatment on M2 high speed steel tools after using either laboratories or shop oor tests in an automotive industry. Sliding abrasion and hardness tests were also carried out as well as microstructural analysis. Advantages were found for the treated tools in some of these tests. 2006 Elsevier B.V. All rights reserved.
Keywords: High speed steel tools; Cryogenic treatment; Improving machinability; Wear behaviour

1. Introduction The need to develop more and more resistant tool materials able to cut increasingly resistant workpiece materials and the demand of the technological development for higher productivity and lower costs have caused many tool materials with excellent properties such as cemented carbide, cermets, ceramics and ultra hard materials (CBN, PCBN and PCD) to emerge. High speed steel (HSS) can also be included in this list since this tool material is fairly well used in the industry to date, although being developed more than a century ago. Its main applications are for drills, taps, milling cutters, broaches and also bits where the economical cutting speed is too low to think about carbide tools [1].

Corresponding author. Tel.: +55 34 3239 4186; fax: +55 34 3239 4272. E-mail addresses: avio@ct.ufes.br (F.J. da Silva), ltm-sdfranco@ufu.br (S.D. Franco), alissonm@mecanica.ufu.br (A.R. Machado), ezugwueo@lsbu.ac.uk (E.O. Ezugwu), antonio.maria@br.pwtech.com (A.M. Souza Jr.). 0043-1648/$ see front matter 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.wear.2006.01.017

With the time, since their creation, the properties of the HSS tools were considerably improved. The perfect combination of alloying elements and the domain of heat treatment processes conferred to this material excellent hardness and wear resistance properties allied to good toughness. As a result of technological advance a great variety of HSS tools are actually available, including coated and powder metallurgy tools. In recent decades interest in low temperature effects have been demonstrated particularly during heat treating cycles of tool steels. Some literature data indicates that the lives of tools and other steel components may increase signicantly after being submitted to subzero (below 0 C) temperatures. The results can be surprisingly good, depending on the application. Reports of 92817% increases in tool lives after they have being treated at 196 C are found [2]. Unlike coatings that are only a supercial treatment, the cryogenic treatment is applied to the whole volume of the material, reaching the core of the tools. This guarantees maintenance of their properties even after regrinding or resharpening. However,

F.J. da Silva et al. / Wear 261 (2006) 674685

675

the lack of common sense in the literature regarding to the metallurgical aspects that cryogenic treatment confers better wear resistance and consequently higher tool lives as well as contradictory results that are also encountered [35] lead to many doubts and questions involving the practical application of this sort of treatment. The articles found in the literature about this subject vary tremendously from mere promotional information until full scientic publication. The latter kind of articles may offer metallurgical details and the mechanisms that will guarantee improved properties of the tools. The rst users of this technique [6] applied temperatures in the range of 80 to 100 C for periods of about 30 min1 h, and the improvement on tool life was credited to the transformation of retained austenite (softer) into martensite (harder) and the production of a more stable structure. In general the addition of alloying elements lowers the Ms (temperature of the beginning of martensite transformation) and Mf (nal transformation temperature) lines in a way that the latter dwells at subzero temperatures. The conventional heat treatment normally uses cooling conditions only until room temperature, which may leave some retained austenite on the microstructure. This fact must be considered during heat treatment of tool steels. For the eutectoid steel the Mf temperature is of approximately of 50 C, therefore after quenching some percentage of retained austenite will be present [8]. Lately this structure can be transformed into martensite if the material is submitted to reheating or to a stress eld, causing distortion on its body. This non-tempered martensite may cause cracks, particularly in complex shape tools made of highly alloyed steels [9]. The subzero treatment will transform a great deal of this retained austenite by reaching the Mf line, giving more dimensional stability in the tool microstructure. Barron [10] has attributed the improvement of the wear resistance of these tools to another mechanism besides the transformation of the retained austenite into martensite. He veried that the tool steels submitted to conventional heat treatment presented only a small amount of retained austenite, but those submitted to cryogenic treatment showed better performance during machining. This new mechanism would be time and temperature dependent due to the long period (8 h or more) during which the tools would have to stay at cryogenic temperatures. Before the cryogenic treatment the microstructure showed relatively large carbides (20 m) dispersed in the matrix. After the cryogenic treatment, carbide particles as small as 5 m were found. The carbide renement could in such a way contribute to the improvement of the wear resistance of the tool. Barron thus attributed this achievement both to austenite transformation and to the presence of hard and small carbide particles well distributed among the larger carbide particles within the martensite matrix. Popandopulo and Zhukova [11] carried out dilatometry studies and microstructure analysis during cryogenic treatment. They observed volume reduction of the specimen at the temperature range of 90 to +20 C. This behaviour was attributed to partial decomposition of the martensite and precipitation of

Table 1 Various kinds of heat treatment studied by Alexandru et al. [13] A B C G H M N Quenching 1230 C Quenching 1230 C + double tempering 560 C Quenching 1230 C + sub-zero (70 C) Quenching 1230 C + sub-zero (70 C) + tempering 560 C Quenching 1230 C + tempering 560 C + sub-zero (70 C) Quenching 1230 C + tempering 560 C + sub-zero (70 C) + tempering 560 C Quenching 1230 C + tempering 560 C + sub-zero (70 C) + tempering 560 C + sub-zero (70 C) + tempering 560 C A (%) 42.6 12.8 7.5 2.2 13.3 1.6 0.9 M (%) 66.6 74.6 84.9 85.3 71.2 19.9 81.7 C (%) 6.9 12.5 7.6 12.5 15.5 18.5 17.4 NC 23410.24 30928.49 23788.52 42869.81 69646.09 T (min) 22 38 49 47 51 45

Route A B C G H M N

A: austenite; M: martensite; C: carbides; NC : amount of carbides smaller than 1 m/mm; T: tool life.

carbon atoms at dislocation lines and formation of ultramicroscopic carbides. Paulin [2] also veried the presence of ne precipitated carbide particles and their importance to the material properties. The precipitated carbides reduce internal tension of the martensite and minimize microcracks susceptibility, while the uniform distribution of ne carbides of high hardness enhances the wear resistance. Huang et al. [12] conrmed that cryogenic treatment not only facilitate the carbide formation but can also make the carbide distribution more homogeneous. The main variables during heat treatment have a great deal of inuence on the results. A research done in steels equivalent to M2, varying the cryogenic cycles has quantied the precipitated particles and veried their inuence onto the material properties [13]. Their research involved seven steel samples, each of them submitted to different heating and cooling (up to 70 C) cycles as described in Table 1. The microstructure was analysed and the carbide particles quantied using SEM, X-ray difractometer, quantitative metallography and differential dilatometer. The results conrmed an increase in carbide precipitation (from 6.9% to 17.4%), a reduction of the retained austenite (from 42.6% to 0.9%) and an increase in the martensite content (from 66% to 81.7%). The machining tests carried out with bits in turning AISI 1050 steels showed a signicant increase in tool lives of cryogenically treated tools (Table 1). These results can be attributed to minimum quantity of retained austenite, higher amount of martensite content, higher density of ne carbides (smaller than 1 m) and a more favourable distribution of the alloying elements among the carbide of the matrix. Barron [14] after cryogenically treating several materials including the M2 high speed steel at 84 C (maintaining it at this temperature for 24 h) observed a signicant improvement on the wear resistance in sliding abrasion tests [15] when compared to conventionally heat treated steel (quenched and tempered).

676 Table 2 Different cycles applied to M2 high speed steel [16] A B C D E

F.J. da Silva et al. / Wear 261 (2006) 674685

Quenching from 1250 C + triple tempering at 560 C Quenching from 1250 C + 1 cycle 24 h sub-zero at 196 C + triple tempering at 560 C Quenching from 1250 C + 1 cycle 48 h sub-zero at 196 C + triple tempering at 560 C Quenching from 1250 C + 3 cycle totaling 48 h sub-zero at 196 C + triple tempering at 560 C Quenching from 1250 C + triple tempering at 560 C+ 1 cycle 48 h sub-zero at 196 C Hardness (HRC) 63.7 64.8 65.0 65.4 64.3 Red hardness (HRC) 600 C 60.6 62.1 63.0 63.1 61.8 625 C 57.8 57.1 59.3 61.7 58.1 630 C 55.7 57.5 58.0 59.5 57.3 2583 2880 2873 3096 2611 3.5 4.4 4.4 5.0 3.9 BS (MPa) IT (J/mm2 )

treated tools with the conventionally treated ones, highlighting the tests carried out in the laboratory and at the shop oor of a car manufacturer industry. 2.1. Cutting tools The cutting tools used in the machining tests are: A and D: Lathe tool of M2 high speed steel with the dimensions of 10 mm 10 mm 102 mm. B and E: Twist Drills of M2 high speed steel with 7.5 mm of diameter. C and F: Special milling cutter of M2 high speed steel with a 3 m TiN coating. The rst two tools (A and B) were laboratory tested while the last tool (C) was tested at the shop oor of a car manufacturing industry. The tools came from the same batch to avoid possible performance variation due to variation caused by the manufacturing process. 2.2. Cryogenic treatment The tools were cryogenically treated at Cryo Quality Ltd. Company using equipment that could completely control the thermal cycle in terms of temperature and time. A recommended thermal cycle for this tool material was used, consisting of a cooling to a temperature of 196 C followed by three cycles of heating to temperatures in the order of +196 C for tempering, lasting a total of 43 h. Fig. 2 illustrates this thermal cycle. The following steps were taken for the cryogenic treatment, after the tools were conventionally quenched and tempered: Step 1: Cooling to 196 C (4 h at a rate of 1 C/mim); Step 2: Cold stabilization at 196 C (20 h); Step 3: Heating to +196 C (8 h at a rate of 1 C/mim); Step 4: Hot stabilization at +196 C (2 h); Step 5: Cooling to room temperature (1 h average); Step 6: Stabilization at room temperature (2 h); Step 7: Heating to +196 C (1 h average).

Route

A B C D E

BS: bending strength; IT: impact toughness.

When the temperature of the cryogenic treatment was reduced further to 196 C, the wear resistance was increased even more. In a more recent work Yun et al. [16] veried changes in the microstructure of M2 high speed steel when this material was submitted to different cycles of cryogenic treatment at 196 C (Table 2). Comparing the conventional quenching cycle A with other cryogenic cycles it was observed increases of 11.5% in the bending strength, 43% in the toughness and changes in the room temperature and hot hardness. The results were also attributed to transformation of the retained austenite into martensite and precipitation of ultra-ne carbides, with this latter being considered the key point for the changes in the properties. Tests with M2 steel milling cutters after cryogenic treatment (cycle D) in the machining of piston rings of grey cast iron showed a production of 440 rings against 220 rings produced with the conventional heat treated tools (cycle A). It is common sense that transformation of the retained austenite and precipitation of micro-carbide particles are the main mechanism responsible for the better wear resistance of the cryogenically treated tools. However, possibly other mechanisms can also be present since positive results were also veried after cryogenic treatment of copper alloys (welding electrodes), cemented carbides and aluminium alloys for aerospace industry [7]. A better understanding of these mechanisms and how they can affect the properties of the material become mandatory, principally when these properties will directly inuence the performance of the tools. The literature is always showing benets of the cryogenic heat treatment of cutting tools, moulds and dies [1722]. The present work experimentally studies the performance of cryogenically treated high speed steel tools comparing with tools of the same material but conventionally heat treated, during machining and during sliding abrasion tests. 2. Experimental procedure Fig. 1 shows a owchart with the resume of the activities developed in this work in order to compare the cryogenically

Steps 57 were repeated three times. Before the cryogenic treatment the tools have previously been submitted to conventional thermal treatment to obtain the secondary hardness (conventional quenching and tempering). This sequence was chosen following the work developed by Yun et al. [16]. According to them the cryogenic treatment in M2 high speed steel can be applied either after quenching and tempering or straight after the quenching. Their results with tools cryogenically treated straight after the quenching were apparently better than those obtained with the tool cryogenically treated after quenching and tempering. Regardless the rout the materials can usually have their properties improved with cryogenic treatment.

F.J. da Silva et al. / Wear 261 (2006) 674685

677

Fig. 1. Experimental tests carried out in this work.

2.3. Laboratory tests 2.3.1. Lathe toolsbits 2.3.1.1. Microstrutural analysis and hardness. This part of the work had the objective of analysing the changes that occurred in the microstructure of the M2 steel after the cryogenic treatment.

The percentages of the retained austenite for both treated and untreated steels were determined according to the ASTM E 975-03 [23] standard, using an X-ray difractometer. The content of carbide not dissolved during austenitisation was taken into consideration in determining the amount of retained austenite. The hardness (HRc) was measured on three cryogenically treated samples and on three non-treated samples with a minimum of four indentations in each. The Vickers HV0.1 microhardness was measured on one sample cryogenically treated and on one sample non-treated with a minimum of ten indentations in each. The average of these measurements was considered for comparison. 2.3.1.2. Brandsma (1936) rapid facing tests. This is a short lasting test based on tool life which consists of face turning a disc from the center towards the periphery with spindle rotation and feed rate constants as depicted in Fig. 3. It starts with a small cutting speed vc1 , corresponding to the diameter of the hole at the center of the disk reaching a much higher cutting speed vc2 at the end of the test, dened by a complete destruction of the cutting edge [24].

Fig. 2. Processing routes of the cryogenic treatment.

678

F.J. da Silva et al. / Wear 261 (2006) 674685

Fig. 3. Detail of the Brandsma rapid facing test.

In Fig. 3, vc1 is the initial cutting speed (m/min), vc2 the nal cutting speed identied when the cutting edge is completely destroyed (m/min), d1 and d2 the initial and nal diameter of the disk, respectively (mm) and n is the spindle rotation (rpm). As a comparative parameter of this method the diameter machined by the tool up to the end of its life (complete destruction), d2 was used. The end of the tool life was very conspicuously identied because the surface roughness was visible deteriorated straight after the destruction of the cutting edge. The M2 lathe tools (square section of 10 mm 10 mm) were ground to the following geometry: r = 75 ; r = 90 ; s = 0 ; o = 6 ; o = 10 . The tests were carried out on an IMOR MAXII-520 lathe with six CV of power. A 25.4 mm thick AISI 1020 steel disk with d1 = 30 mm and external diameter (maximum d2 ) of 300 mm. These dimensions were sufcient to guarantee tool destruction with the cutting conditions used in all tests. The depth of cut was kept constant and equal to 1 mm for all tests. In a rst round of the tests the feed rate was xed in 0.069 mm/rev and the spindle rotation varied (140, 180 and 224 rpm). In the second round the spindle rotation was kept constant and equal to 140 rpm and the feed rate varied (0.069, 0.109 and 0.157 mm/rev). At the end of each test the disk was completely cleaned by facing it with a sacricing tool. Only after this operation a new test with a new cutting tool were initiated. 2.4. Slinding abrasion test In the present investigation small cylinders of M2 high speed steel with 4.9 mm of diameter and 20 mm long with and without cryogenic treatment were used. These cylinders were taken from square section bars by electrical discharge machining (EDM) which were then turned in a lathe with ceramic tools for diameter corrections. The abrasive tests were carried out in a multiple pin-on-disc abrasion tester. In this kind of apparatus the disk rotates and at the same time the rotating pin is fed, describing an Arquimedes spiral (Fig. 4). This guarantees that always a fresh abrasive part of the disc is in action because the pin never passes on the same place again. The pin and the disk are rotated in the clockwise direction at 25 rpm and the average relative velocity of the pin against the disc was 110 mm/s. The test length was 3220 mm. In the rst part of the tests a 600 mesh alumina abrasive paper (Al2 O3 -Norton T223) was used and in the second part the grain

size of the abrasive were varied to 80 mesh. The normal load applied was of 10 N. At least ve pre-tests were carried out in order to stabilize the wear rate. After this stabilization a minimum of ve tests for each sample and condition were performed. The mass loss was determined by an analytical scale with 104 g of resolution, and the wear rate was calculated dividing the mass loss by the area of the pin and by the test length. 2.4.1. Twist drills The drilling tests were carried out in a ROMIInteract IV machine center equipped with a CNC MACH 4, with 16 kW of power and variable speed (404000 rpm) and feed (04800 m/min). AISI 8640 steel bars with an average hardness HV of 290 (kgf/mm2 ), having a square section of 110 mm 500 mm long were used as work material. Cryogenically treated and untreated SAE M2 HSS twist drills with 7 mm of diameter, manufactured by Twill S/A were used in the drilling tests. During machining power consumption was measured by a Hall sensor type NW-SC-50. The sensor actually measures the electric current that runs the main spindle of the machine tool and with the aid of a CYRDAS 1602 data board and a microcomputer the signal is processed and the power consumption is calculated and directly displayed. The power consumption during cut is determined by the reading made during drilling diminished by the reading made with the machine running freely. Tool wear was measured with the help of a table optical microscopic equipped with a Mitutoyo gauge with a resolution of 0.01 mm. At the end of the tool life the rake and ank faces of the drills were analysed within a SEM.

Fig. 4. Schematic diagram of the pin-on-disc test for sliding abrasion.

F.J. da Silva et al. / Wear 261 (2006) 674685

679

Fig. 5. Direction of application of the cutting uid during drilling.

An emulsion of mineral oil at 5% of concentration was used as cutting uid applied at a rate of 5 l/min according to the illustration shown in Fig. 5. The cutting speeds used were 30, 35 and 40 m/min and the feed rate was constant at 0.11 mm/rev. A 3L/D (length of the role/diameter of the twist drill) rate was always used where the hole length was then of 22.5 mm. The twist drills were ground with the following geometry: r = 67.5 ; r = 112.5 ; r = 0 . The end of tool life criterion was the catastrophic failure, recommended by NORDTEST NT MECHE 038 [25] standard. According to this standard the tool life is determined by the number of holes machined until immediately before tool failure. The maximum and medium ank wear VBB max and VBB were frequently measured in order to have continuous control of the behaviour of the wear development. This allowed to determine the tool life under different criteria, e.g. VBB = 0.3 mm and VBB max = 0.7 mm. The frequency of tool wear measurement was determined by the severity of the test. After eight holes machined the tool wear was measured when the cutting speed was 30 m/min and after four holes machined when the cutting speed was 35 and 40 m/min. 2.5. Shop oor tests The shop oor tests were carried out in a car manufacturer plant. Special shaper milling cutter was used in these experiments. Both cryogenically treated and untreated tools were used under industrial cutting conditions. The cutters were also SAE M2 HSS, this time TiN coated. They have two teeth and were used in a HURTH ZK7 machine tool that manufactures the dented ring that engages the rst and second speed in the gear box. During machining two shaper milling cutters are used to generate the right and the left surface of the top tip of the teeth of

the dented ring shown in Fig. 6. In this operation a straight line is produced at the mid top of each tooth which has the function of facilitating engagement of the gears. The dented ring is made of 19MnCr5G according to DIN 17006 standard. The right and the left milling cutters operate simultaneously at different tooth in such a way that at the end of the cycle all the teeth are shaped. Tool wear was monitored with the help of an optical microscope. The roughness of the generated surface was measured by a PERTHEN-58p meter, with a cut-off of 0.25 mm. At the end of tool lives the tools were analysed within a SEM. All the cuts were performed dry with a cutting speed of 89 m/min and a feed velocity of 1200 mm/min. New and resharpened tools were tested. It is worth to mention that after resharpement the tools lose the coating at their main ank face. The tool life was considered ended after 200 workpieces had been machined with the new tools and after 150 workpieces with the resharpened tools. This end of tool life criteria was determined based on shop oor experience. Average ank wear and the surface roughness were monitored at the beginning and towards the end of the tool lives. In a second round a different end of tool life criterion based on the observation of burr was adopted. This is an important criterion since the workpieces are not submitted to a deburring process after machining and they would compromise the quality of the dented rings. Fig. 7 shows on the B detail the usual location of the burr. The inspection was done visually and in order to minimise subjectivity errors only one person took charge of these tests. 3. Results and discussion 3.1. Laboratory tests (lathe tools) 3.1.1. Microstrutural analysis The X-ray difractometer results showed a difference in retained austenite between the cryogenically treated and untreated tool samples. The untreated tool sample showed a volume of approximately 25% of retained austenite against nearly 0% for the cryogenically treated tool sample. Table 3 shows the hardness of these samples. They are practically the same; therefore the cryogenic treatment had no inuence on this property of the tools.

Fig. 6. View of the top of the teeth illustrating the surfaces generated by the right and the left shaper milling cutters.

680

F.J. da Silva et al. / Wear 261 (2006) 674685

Fig. 7. Burr observed on a workpiece at the end of the tool life.

Microhardness results also did not show conspicuous difference between the treated and untreated tools. The treated samples had a mean value of 820 HV0.1 against 819 HV0.1 for the untreated tools. Barron [10] also did not nd signicant difference between the cryogenically treated and untreated M2 tools. According to Collins [26] the precipitation of ne carbides during the cryogenic treatment cycle causes an increase in the wear resistance and in the tool toughness, but only a small, if any, in the tool hardness. Actually he observed that initially the hardness falls sharply at the cryogenic cycle and when the tool is heated to the room temperature the hardness is totally recovered. 3.1.2. Brandsma (1936) rapid facing tests Fig. 8 shows the results of the rapid facing tests. It is noticed that increasing both the rotate speed and the feed rate usually reduce the tool lives regardless the heat treatment submitted by the tool. The cryogenically treated tools always presented longer tool lives when the spindle speed was varied. The results indicated increases of 12.5%, 44.0% and 21.0%, respectively, for tests under 140, 180 and 224 rpm. When the feed rate was varied the treated tools also presented better performance under the

Fig. 8. Rapid facing test results.

Table 3 Effect of the cryogenic treatment on the tool hardness Treated tools A B C HRc 66 65 66 Untreated tools D E F HRc 66 65 66

two smallest feed rates and only with the highest feed rate of 0.157 mm/rev the untreated tool presented a marginal better tool life than the treated tools (diameter of failure of 94 mm for the untreated tool against 91 mm for the cryogenically treated tool). These results suggest that the cryogenically treated tools are generally better than the untreated tools but the cutting conditions have great inuence on their performances. The transformation of almost all retained austenite into martensite, a harder structure, and precipitation of ne and hard carbides as seen in the literature, are surely responsible for these positive results of the cryogenically treated tools. This usual superior performance can only be beaten when the relatively high percentage of the soft retained austenite of the untreated tools transforms into hard martensite during the severe action of the cutting operation giving them a good wear resistance. This might explain the better performance of the untreated tool when compared to the performance of the cryogenically treated tool at the highest feed rate of 0.157 mm/rev.

F.J. da Silva et al. / Wear 261 (2006) 674685

681

Fig. 9. Wear rate of the pin-on-disc for sliding abrasion tests (Al2 O3 abrasive paper).

3.1.3. Sliding abrasion test Fig. 9 presents the average wear rate obtained after sliding abrasion wear testing using the cryogenically treated and untreated samples with 600 and 80 mesh abrasive papers. With increasing the abrasive grain size the wear rate increases proportionally as expected since the abrasive diameter change from 15 m (600 mesh) to 180 m (80 mesh). Statistical analysis of the variance for the tests with 600 mesh for both cryogenically treated and untreated samples did not show signicant difference of the average wear rates with 95% of reliability. However, for the 80 mesh the average wear rate for the cryogenically treated sample was slightly smaller (3.3%) than the untreated sample. The transformation of the approximately 25% of the retained austenite into martensite after cryogenic treatment did not lead to a signicant alteration of the abrasive wear rate at the conditions used in the pin-on-disc tests. Literature results [27] showed that, depending on the test parameters such as normal load, average grain size and type of the abrasive, quantity and shape of the carbides, among others, the increasing of the amount of retained austenite can lead to an increase or to a decrease in the wear rate of ferrous alloys. This behaviour is associated to the ability

of the austenite to harden during plastic deformation either by workhardening or by martensite transformation. The abrasive wear tests carried out by Barron [14] on the M2 HSS cryogenically treated at 196 C (cooling rate of 3 C/mim and maintained for 24 h) had an increase on the wear resistance of 20% when compared to untreated samples. Discrepancies of Barrons results with those of the present investigation may be credited to different parameter conditions used. At Barrons tests the normal load was 430 N, feed velocity of 0.48 m/s and sample diameter of 12.7 mm against 10 N, 110 m/s and 4.9 mm, respectively, at the present investigation. This is really a test very sensible to the test conditions. The wear resistance as a function of the ne carbides that has possibly precipitated during cryogenic treatment is also very dependent on the test conditions chiey on the feed velocity and percentage of precipitation of carbides [28]. The worn surfaces of the cryogenically treated and untreated samples are shown in Fig. 10. It is observed that the wear mechanisms were microplowing and microcutting. No significant difference can be noticed regarding to the proportion of these two mechanisms, suggesting, therefore, an effective workhardening process or even transformation of the austenite into martensite during test of the untreated sample, as veried by Zum Gahr [27] in several ferrous materials and SiC abrasive. The hardness and the abrasive tests did not show results that could justify the better performance of the cryogenically treated HSS tools in machinability tests. During machining the phenomena are very specics, with very high chiptool interface temperatures, very high tension on the tool surfaces and great interaction between the tool and the work material and compared to sliding abrasive tests the wear occurs in a smaller scale. At the cutting conditions and tools (HSS) used in the present investigation the main wear mechanisms expected is abrasion and attrition [29]. The results obtained here showed that probably the transformation of the retained austenite into martensite and precipitation of ne carbides during cryogenic treatment is benecial in the presence of such wear mechanisms.

Fig. 10. Typical worn surface of (a) treated and (b) untreated samples, after pin-on-disc tests with 80 mesh Al2 O3 abrasive paper.

682

F.J. da Silva et al. / Wear 261 (2006) 674685 Table 4 Number of holes for end of tool life criteria VBB = 0.3 mm, VBB max = 0.7 mm and catastrophic failure 30 m/min 35 m/min 40 m/min

Untreated Treated Untreated Treated Untreated Treated VBB 0.30 mm VBB max 0.70 mm Catastrophic failure
a

120
a

a a

12
a

26
a

208 368

88 91

149

55

14

62

Criterion not reached.

Fig. 11. Twist drill lives given in numbers of holes until the catastrophic failure.

3.2. Laboratory tests (twist dills) Fig. 11 summarizes the results of tool life tests given in numbers of holes machined. It can be observed that increasing the cutting speed the lives of the twist drills reduces considerably regardless the heat treatment condition of the tool material. This is due to the increase of the chip tool interface temperature with increasing cutting speed. The lives of the cryogenically treated tools were always superior to those given by the untreated tools. In percentage terms the differences were 147%, 65%, 343% for the cutting speeds of 30, 35 and 40 m/min, respectively. When other end of tool life criteria is used, according to Table 4, different results are produced. Considering the end of tool life criteria of VBB , VBB max and catastrophic failure simultaneously lives of the cryogenically treated and untreated tools is modied to values presented in the curve of Fig. 12. Although with lower percentages the cryogenically treated tools performed far better than the untreated tools. The percentages are now modied to 73%, 60%, 117%, respectively, depending on the cutting speed used.

Fig. 12. Number of holes machined for simultaneous end of tool life criteria of VBB = 0.3 mm VBB max = 0.7 mm and catastrophic failure.

Fig. 13 shows the worn areas of the ank face of the twist drills used with the cutting speed of 40 m/min. The wear of the untreated tool, after 14 holes machined is clearly bigger than the cryogenically treated tool, after 62 holes machined. The power consumption when drilling with both cryogenically treated and untreated tools were generally very similar and only occasionally the untreated tools showed lower power consumption than the treated tools during drilling. This could be an indication that the microstructure changes (transformation of

Fig. 13. Worn areas of the ank face of the twist drills used at vc = 40 m/min and f = 0.11 mm/rev: (a) with cryogenic treatment (after 62 holes) and (b) without cryogenic treatment (after 14 holes).

F.J. da Silva et al. / Wear 261 (2006) 674685

683

the retained austenite into martensite and precipitation of ne carbides) do not cause modication on the chip formation process to the point of modifying the cutting forces, but they can enhance wear resistance as shown previously. However, this analysis needs further subsidies because when drilling four different work materials with cryogenically treated and untreated T1 twist drills at various cutting conditions, Cohen and Kamody [30] found lower power consumption for the treated tools for three work materials (33%, 8.5% and 4.7%) when compared to the untreated tools and only for one work material the untreated tool showed smaller (6%) power consumption than its treated counterpartner. Microstructure changes may have contributed to Cohen and Kamodys results. 3.3. Shop oor tests (special shaper milling cutter) Fig. 14 presents the values of the average ank wear of the right and left milling cutters (new tools) after machining 200 workpieces for the teeth nos. 1 and 2. The cryogenically treated tools present higher average ank wear than the untreated tools. This fact was also observed when using resharpened tools. Although the wear of the cryogenically treated tools were higher than the untreated ones the surface roughness generated by the two tools were very similar (Ra varying from 0.99 to
Fig. 14. Flank wear of milling cutters after machining 200 workpieces.

1.63 m for the surface generated by the treated new tools and from 0.98 to 1.41 m for the new untreated tools). Fig. 15 shows the cutting edges of the right hand side milling cutters of the cryogenically treated and untreated new tools after machining 200 workpieces. It is clearly seen the higher wear developed on the cryogenically treated tools compared to the untreated ones. It is noticed that the coatings at the ank face of the tools were worn out.

Fig. 15. Right milling cuttersnew after 200 parts produced.

684

F.J. da Silva et al. / Wear 261 (2006) 674685

developed during the thermal treatment at the coatingsubstrate interface that could help to explain the process of deterioration of this kind of tools. 4. Conclusions Based on the results obtained in the present investigation the following conclusions can be drawn: 1. The hardness and the microhardness of the M2 HSS samples were not signicantly affected by the cryogenic treatment. 2. The samples cryogenically treated showed a fraction very close to 0% of retained austenite. This means that practically the 25% in volume of the retained austenite observed in the untreated sample were transformed into martensite by the cryogenic treatment. 3. A superior performance of the cryogenically treated tools compared to the untreated ones was observed in the Brandsma rapid facing test. This difference reached 44% in some cutting conditions. 4. The difference on the percentage of retained austenite of the cryogenically treated and untreated samples did not alter the abrasive wear rate at the conditions used here for the sliding abrasion tests. This is possibly due to the ability of the austenite of the untreated samples to harden during plastic deformation either by workhardening or by its transformation into martensite. These phenomena may compensate the gain obtained by precipitation of ne carbides in the cryogenically treated samples. 5. The cryogenic treatment increased the performance of the M2 HSS twist drills. The gain observed during drilling steels adopting catastrophic failure as the end of tool life criterion varied from 65% to 343% depending on the cutting conditions used. 6. Shop oor tests with cryogenically treated coated HSS milling cutters presented worse performance than untreated tools when shaping the top surface of the teeth of gear rings. 7. Overall the cryogenic treatment had favourable inuences on the performance of the tools tested. This means that depending on the application the cryogenic treatment may be a good alternative for having productivity enhancement. Optimization of the parameters involved in the whole thermal cycle must, however, precede the application. Acknowledgements The authors would like to thank Cryo Quality Ltd. Company for carrying out the cryogenic treatment in the samples; CAPES for the scholarship offered to one of the authors; FAPEMIG for the nancial support and Fiat-GM PowerTrain for technical support. References
[1] A.R. Machado, M.B. da Silva, Usinagem dos Metais (Metal Machining), 3rd ed., Editora da Universidade Federal de Uberl andia, MG, Brazil, 2003, p. 275 (in Portuguese).

Fig. 16. Average of parts produced by the cryogenically treated and untreated tools when using the appearance of burrs as the end of tool life criterion.

Fig. 16 shows the production of the cryogenically treated and untreated tools when using the formation of burrs as the end of tool life criterion. In these tests the tools were resharpened four times, resulting in a total of 380 parts produced by the cryogenically treated tools and 491 parts produced by the untreated tools with an average of 95 and 123 parts produced per life by the cryogenically treated and untreated tools, respectively. It was seen that the cryogenic treatment in such a way had a negative effect on the performance of the tools when compared to the untreated ones. According to Paulin [2] the cryogenic treatment can be applied to coated tools satisfactorily. The same is cited by Cohen and Kamody [30] who found 42% increase in tool life of cryogenically treated TiCN coated M4 tools when comparing to untreated coated tools in a broaching operation. They did not say, however, whether the cryogenic treatment was applied to the tool before or after the coating operation of the tools. If the treatment was applied after the coating operation a critical tension state can appear between the coating and the substrate causing adherence problems. These materials have different expansion coefcient, being 12 106 K1 for the HSS and 9.4 106 K1 for the TiN. This can create triaxial tension state at the coating and substrate interface that can promote the fragmentation and rapid wear of the coating layer. The appearance of burrs is directly related to the development of the wear and the results shown here reects the relative low wear resistance of the cryogenically treated tools compared to the untreated counterpartners as seen. The results presented by this investigation reveal that the cryogenic treatment is a technique that demands further studies not only on uncoated tools but also on coated tools. It is known that there is transformation of austenite into martensite and precipitation of ne carbides that are well distributed into the whole matrix of the cryogenically treated steels [13,16]. The result of the cryogenic treatment and consequently the efciency of the treated tool is very dependent of the thermal cycle, including the cryogenic temperatures involved. The optimisation of this cycle for each individual application must therefore be hunted in order to have success of the technique. The particular case of coated tools demands detailed understanding of the phenomena

F.J. da Silva et al. / Wear 261 (2006) 674685 [2] P. Paulin, Frozen gears, Gear Technol. (1993) 2628. [3] E.A. Smolnikov, G.A. Kossovich, Cold Treatment of Cutting Tools (translated from Metallovedenie i Termicheskaya Obrabotka Metallov), no. 10, 1980, pp. 57. [4] L.B. Tseitlin, V.D. Kolensnichenko, T.V. Karnaushenko, V.V. Umanets, E.S. Zhmud, Tool Life of High Speed Steel Cutters After Cold Treatment (translated from Metallovedenie i Termicheskaya Obrabotka Metallov), no. 10, 1980, pp. 79. [5] E.S. Zhmud, Improved Tool Life After Shock Cooling (translated from Metallovedenie i Termicheskaya Obrabotka Metallov), no.10, 1980, pp. 35. [6] A.P. Gulyaev, Improved methods of heat treating high speed steels to improve the cutting properties, Metallurgy (12) (1937) 65. [7] R.B. Reasbeck, Improved tool life by the cryotough treatment, Metallurgia (1989) 178179. [8] D.S. Zamborsky, Control of distortion in tools steels, in: The Heat Treating Source Book, ASM, 1986, pp. 7379. [9] J.M. Heberling, Tool steel tutorial, Heat Treating (1992) 2230. [10] F.R. Barron, YesCryogenic Treatments Can Save You Money! Heres Why, Tapi 57 (5) (1974) 3540. [11] A.N. Popandopulo, L.T. Zhukova, Transformations in High Speed Steels During Cold Treatment (translated from Metallovedenie i Termicheskaya Obrabotka Metallov), no. 10, 1980, pp. 911. [12] J.Y. Huang, Y.T. Zhu, X.Z. Liao, I.J. Beyerlein, M.A. Bourke, T.E. Mitchell, Microstructure of cryogenic treated M2 tool steel, Mater. Sci. Eng. A399 (2003) 241244. ` [13] Alexandru, G. Ailincai, C. Baciu, Inuence de traitements thermiques a ` outils a ` coupe reapide tr` basse temp eature sur la dur ee de vie des aciers a es, M emoires et etudes scientiques revue de Metallurgie (1990) 283388 (in French). [14] R.F. Barron, Cryogenic treatment of metals to improve wear resistance, Cryogenics (1982) 409413. [15] M.A. Moore, A review of two-boby abrasive wear, Wear 27 (1974) 117. [16] D. Yun, L. Xiaoping, X. Hongshen, Deep cryogenic treatment of high-speed steel and its mechanism, Heat Treat. Met. (1998) 5559.

685

[17] A. Molinari, M. Pellizzari, S. Gialanella, G. Straffelini, K.H. Stiasny, Effect of deep cryogenic treatment on the mechanical properties of tool steels, J. Mater. Process. Technol. 118 (2001) 350355. [18] D. Mohan Lal, S. Renganarayanan, A. Kalanidhi, Cryogenic treatment to augment wear resistance of tool and die steels, Cryogenics 41 (2001) 149155. [19] M.C. Huang, C.H. Gao, L.G. Huang, Study on cryogenic phase change and wear characteristic of high speed steel, Acta Metall. Sin. (English Lett.) 16 (6) (2003) 524530. [20] V. Leskovsek, B. Ule, Inuence of deep cryogenic treatment on microstructure, mechanical properties and dimensional changes of vacuum heattreated high-speed steel, Heat Treat. Met. (UK) 29 (3) (2002) 7276. [21] R. Mahmudi, H.M. Ghasemi, H.R. Faradji, Effect of cryogenic treatments on the mechanical properties and wear behaviour of high-speed steel M2, Heat Treat. Met. (UK) 27 (3) (2000) 6972. [22] X. Lin, Y. Dong, Y. Wang, Study on cryogenic treatment technology and mechanism of high speed steels, Trans. Met. Heat Treat. (China) 19 (2) (1998) 2125. [23] ASMT E 975-03, Standard practice for X-ray determination of retained austenite in steel with near random crystallographic orientation, Book of Standards, vol. 03.01, ASTM International. [24] D. Ferraresi, Fundamentos da Usinagem dos Metais (Fundaments of Metal Cutting), Editora Edgard Blucher Ltd., 1977, p. 751 (in Portuguese). [25] NODTEST NT MECH 038, Cutting uids for drilling: evaluation by drill life test, Nordtest Method, Proj. 1242-95/2, 1997, ISSN 0283-7196. [26] D.N. Collins, Deep cryogenic treatment of tool steels: a review, Heat Treat. Met. (2) (1996) 4042. [27] K.H. Zum Gahr, Microstructure and Wear of Materials, Elsevier Science Publishers Inc., Amsterdam, 1987, p. 590. [28] J.D. Kamody, Using deep cryogenics to advantage, Adv. Mater. Process. 10 (1998) 215218. [29] P. Wright, E.M. Trent, Metal Cutting, 4th ed., Heinemann Ltd., Butterworths, 2001, p. 273. ISBN 0-7506-1068-9. [30] P. Cohen, D. Kamody, Cryogenics goes deeper, Cutting Tool Eng. l50 (7) (1998) 4650.