STRENGTH PROPERTIES

STRUCTURE OF SURFACE LAYERS AND WEAR RESISTANCE OF QUENCHED 50G STEEL AFTER FINISHING, ANTIFRICTION, NONABRASIVE TREATMENT L. M. Rybakova, L. I. Kuksenova, and Yu. A. Nazarov

UDC 669.14.018.298:620. 178.16

The effect of the finishing, anti~ction, nonabrasive treatment (FANT) on the structure of the surface layer of the 50G steel subjected to quenching with rf heating without any friction is investigated. Ideas are formulated about the mechanism of formation of the surface layer. Structural parameters of the quality of the coatings determined by the method of nondestructive testing of the surface layers are presented, which can be used as criteria for rapid estimation of the conditions of technological surface treatment. The aggregate of tribotechnical properties of parts of cylindrical piston group including the wear resistance of linings, rings, and pistons, the coefficient of friction, breaking-in ability of the cylinder-lining--piston-ring pair, and the compatibility between the contacting metallic materials refer to factors that control the reliability and longevity of the internal-combustion engines. These properties can be estimated under service-, bench-testing-, and laboratory-testing conditions. A comparison between the results of laboratory tests using machines that produce reciprocating sliding with the wear data on linings and piston rings obtained during service [1 and others] showed that they were qualitatively in agreement, which indicates that it is expedient to estimate the tribotechnical properties in laboratory tests. These tests happen to offer unique possibilities for obtaining a separate estimate of the quality of a pair consisting of a cylinder--piston set needed for the substantiation of the approach toward the choice the structural and lubricating materials, the conditions of the technological processing of the surface parts leading to reduced losses by friction and wear. One of the methods of improving the tribotechnical properties of the parts of a cylinder--piston set is the brass plating or a finishing, antifriction, nonabrasive treatment (FANT) of the surface [2-4]. This treatment is also technologically simple, as well as effective in preventing seizure and decreasing friction and wear. However, according to the data of [4], depending upon the service conditions the effect of FANT may not even be manifest which gives us reaso~ to assume that a negative result is possible depending upon many factors, particularly the structural state of the treated material. Such cases were observed in practice during the service of internal-combustion engines. In this paper, based on the results of laboratory tests on the lining--ring friction pair, we have analyzed the changes in the structure of the surface layer of the quenched 50G steel subjected to FANT, their effect on wear resistance, as well the quality of the surface layer. Using a 77MT-1 reciprocating sliding-friction machine (at a pressure of 30 MPa and average velocity of 0.1 m/sec), we tested a pair consisting of a lining (50G steel) and a chromium-plated cast-iron ring (the samples were cut out of productionmodel parts). A self-adjusting sample taken out of a ring was held fixed and, through it, pressure was transmitted to the reciprocating motion of the lining sample located in the bath of M-10G2K lubricant. Two types of linings were tested: after quenching with rf heating (RFH) and after a combined treatment (RFH + FANT). Brass plating* [5, 8] was carried out by using a technique developed under the supervision of V. V. Chepelevskii, which has been described in [6]. After each stage of testing before weighing, the sample was washed in benzene and acetone, and was dried in air. The intensity of wear and linear wear were determined by an analytical method from the loss of weight of a lining sample. The microgeometry of the surface was estimated before and after the tests and the structural changes along the thickness of the surface layers were *Performed by V. V. Kisel at the NIIAT. Institute of Mechanical Engineering, Russian Academy of Sciences. Translated from Metallovedenie i Termicheskaya Obrabotka Metallov, No. 3, pp. 5-9, March, 1993.

0026-0673/93/0304-0129512.50

©1993 Plenum Publishing Corporation

129

/ a ,..~.

o.5 I

C~''-'/~" ~'~'A"~ 0

2

iO "~

eoo c %,

IO "q

~0-I: 0

ZO

~o

b

6e

80

~h

Fig. 1. Tribotechnical properties of the pair: a, c) fragments of profilograms of the surface of a lining before (a) and after (b) friction; 1) RFH, 2) RFH + FANT; b) intensity of wear (Ih) and linear wear (h t) of 50G steel after RFH (1, 3) and after RFH + FANT (2, 4).

.,
0,2

Z#

o,

1~ e

Z

4

6Al.lm,~o a

/

/

O,t

~.~ 6 o ~ e g e b

e

4

61~ Mm

c

Fig. 2. Properties of the structure of the surface layer of a 50G steel before friction: continuous line) quenching with rf heating; broken line) RFH + FANT; chain line) normalizing.

investigated. The structure was investigated by a special technique of a sliding beam of x-rays which, without destroying the surface, enabled us to carry out a layer-by-layer monitoring of the structure of the deformation zone of the metals during friction [7]. The exposures were taken in cobalt K~ radiation, the thickness of the effectively reflecting layers was equal to 0.5-7.5 #m depending on the conditions of the exposure (assuming that 75 % of the radiation was absorbed by the material participating in the reflection). The quality of the deposited layer was estimated by the structural properties [9]. The results of investigations of some tribotechnical properties of the pair after the testing are presented in Fig. 1. It can be seen from Fig. la that the surface of the sample became somewhat smoother after FANT. Considering the conditions of the prior mechanical work of the lining and the state of the material after the FANT, we can assume that the main reasons for the improvement of the microgeometry are not only the filling of depressions by the brass but also the deformation of protrusions during the finishing antifriction treatment. An estimate of the intensity of wear showed that, only after about 75 hours of testing, did the curves of the dependence Ih = f(r) merge into a single one corresponding to the steady-state frictional regime. One should note that, after the rf heating, this regime was attained in a considerably shorter time, approximately after 15 h of testing. In the stage of breaking in, the sample subjected to a combined treatment had a large amount of wear (Fig. lb, curve 4). The wear rate of the samples after the RFH treatment was equal to - 0.01/zm/h and, after the FANT, it changed up to 75 h and, only after this time, one could probably speak of some apparent rate of wear. The surface profiles of all samples were practically identical after testing (Fig. lc). Thus, the results of tests with the friction machines in the laboratory 130

(# ~co,zaya z i

i

i

I

I

I

I

i

5o

',

,, I I

30

I !

I I

a

7O

i;

i

,

l

(e t l) (2zo)

:

(zoo)

50 #0 , I

yo f zo

I I

#0

0

-

gO

o,5

I

,

I

1 I I r

I 1 I I

t~sinZO/,~'~ "z

b

Fig. 3. The dependences/~2cos~0/~k2 = f(KsinZ0/k23,2): a) RFH treatment; b) RFH + FANT. Thickness of the layer: 1) 7.5 t~m, 2) 0.5 /~m, 3) 3.0 tzm, and 4) 1.0 p,m.

/"

a-tee

~W),r

j~aC-fe

8

ac-Fe

=

Fig. 4. Microphotometric traces of the regions of x-ray diffractograms of the 50G steel before friction: a) RFH treatment, b) RFH + FANT. indicated that antifriction, nonabrasive, surface treatment of parts made of 50G steel after inductive quenching increased the time of breaking in, during the stage of which the wear was greater and, hence, the thickness of the worn layer was greater. One should note that, during this process, only the coating was destroyed and the base metal was virtually untouched. The structural properties of the surface layer of the steel before friction determined by the x-ray structural method are presented in Fig. 2. It can be seen that the physical broadening of the x-ray lines (the) after RFH was greater than after normalizing (Fig. 2b). The thickness of the layer analyzed (Fig. 2a)/3(22o) changed by a small amount during the process. After FANT,/3(220) decreased which is basically due to a decrease in the microstresses (as it followed from an analysis of the Hall lines on Fig. 3). After RFH Ad/d (h = 7.5/~m) -- 2.4.10 -3 for D --- 180 A and Ad/d (h = 0.5-3.0/zm) = 2.4.10 -3 for D =- 160 A; after (RFH + FANT) Ad/d (h = 7.5 t~m) = 1.3.10 -3 for D ~ 190 A. Therefore, during frictional deposition of the coatings, thin surface layers of the quenched steel get mechanically heated, which gives rise to recovery of the processes

131

A a', full

+0,002

"Q-.o~

0

Z

-OO , OZ -0,004

',,,%I t.o "o..~/-" /_o.~ 8 I /

#, pt

X m/

-0,o06 Fig. 5. Variation in the nature of the crystal lattice (Aa)e_Fe along the thickness of the surface layer of the steel caused by FANT (h is the distance from the surface): 1) before friction, 2) after friction.

~zz~ l~J,deg

l/3Zcaseg/,l e

(zlO ;{ia) reoa)' e ,4

9L

0

\

30

it~II

Z5

z0] 0

~fIi

/1/ . :i

i

I

I '

/rel

0,2

/

f

.... s - -

Io o Z

<, b

6 h.'p~

l

z V

/I ~,s"

I

-r t I

L

O.5

1,0

c

I

"S¢tsz"2z#°

Fig. 6. Structural properties of the surface layer of the 50G steel after friction: continuous line) RFH treatment; broken line) RFH + FANT; 1: h) 0.5 #m, 2) 7.5/zm, 3) 1.0/zm, 4) 4.5 #m. that typically take place in plastically deformed polycrystals. Also the variation of the relative intensity of the interference lines on the x-ray diffractogram along the thickness of the layers seemed to be systematic (Fig. 2c). From an analysis of the results presented in Fig. 2b, it follows that the presence of the frictional coating on the surface of the steel led to a change in the functional dependence/~(hk.9 = f(0) corresponding to a thin surface layer over the whole range of angles 0 as a result of changes in the nature of the density distribution of the structural defects in the crystallographic planes investigated. On the Hall straight lines (Fig. 3) constructed from the/3(hkt ) values for these layers, a break was observed which indicated that the physical broadening of the x-ray lines increased on account of the contribution of the additional components related to the inhomogeneous chemical composition of the material. Thus, during FANT not only did an intrinsically wear-resistant layer form (which can be clearly seen on the fragments of the x-ray diffractograms of Fig. 4) the thickness of which is equal to a fraction of a micrometer, but a subsurface modified zone - 3 #m in thickness formed as well, the composition of which was different from the matrix. No new phases were de132

tected on the x-ray diffractograms obtained from these layers (except for the intrinsic layer on the sample surface, Fig. 4b), but the lattice parameter of the matrix (Aa = aFANT -- aRFH, Fig. 5, curve I) changed. As in the case of the frictional treatment of the normalized 51213steel [8], the absolute values of Aa and the method of determination of a of the quenched 50G steel indicated that this change could not be due to the zonal stresses (on account of nonuniform surface heating, cooling, and plastic deformation). The main reason was the incorporation of atoms of the friction-rod material into the matrix forming a modified zone. From the experimental data presented above and those published in [8], we can conclude that, after brass plating of the 50G steel, the surface layer consisted of an intrinsic coating of a transitional modified layer and the base metal (matrix). However, in contrast to normalized steel, the surface layers formed on the surface of the 50G steel quenched with RFH had a number of substantial differences: the values of the function/3(~t) = f(t) at all Io.75 were less than after quenching with rf heating; I gradt/3(hk/) [ ItFIt ---- 0.3 and I gradt/3(h~ I RFH+FANT ~ 0.4, i.e., the gradient in the change of density of the crystal defects in the quenched martensite had a thickness of the surface layer less than unity and the ratio between these values were very close to unity for the various forms of treatment; the break on the Hall lines had the opposite sign, i.e., the rays of the broken lines increased (they decreased in the normalized state) [8]; and the absolute values of Aa were approximately three times smaller. The structural differences between the surface layer of the 50G steel after quenching with RFH + FANT heating and after normalizing + FANT were also caused by the differences in the mechanism of the contact interaction between the lining and the ring which were subjected to the above forms of treatment and, hence, tribological effects of the coating. The x-ray diffractograms of thin surface layers after the friction of the samples subjected to various treatments indicated that the phase composition of the deformation zone did not change (the microstructure of the metallic base of the quenched layer is martensite). No residual austenite was detected to exist within the limits of the precision of the x-ray structural analysis (the possibility is not ruled out that it could still be present in an amount less than 5% or that microdiffusion processes occur redistributing the alloying elements inside the matrix, which was manifest in the changes of the c/a ratio). Unexpectedly, it turned out that the lattice parameter varied across the thickness of the layer deformed during the friction (Fig. 5, curve If). It had a pronounced tendency toward decreasing upon approaching the free surface and there was no indication of the presence of any kind of modified subsurface layer present. The disappearance of the effect of the film after a certain length of the frictional path was controlled by the sign, the cause for the change, and the absolute values of Aa, i.e., the decrease in the lattice parameter of the solid solution in the layers adjacent to the interface, which was primarily caused by the adhesive interaction force [9]. The mutual approach between the curves of the change in the relative intensity of the interference lines across the thickness of the deformation zone of the samples subjected to various forms of treatment (Fig. 6b) indicates that the deformation process was the same in the mechanism of the contact deformation which could result from similar properties of the structure in contact with the metal ring (50G steel quenched with rf heating). One might assume that the different natures of the changes in the physical broadening of the x-ray diffraction lines across the thickness of the samples tested (Fig. 6a) were caused by both the different thermal effects on the surface during contact interaction and the differences between the stages of the periodic process of structural changes in the function of the track under frictional-fatigue conditions [10], which is supported by the results of constructing by the Hall method [11] (Fig. 6c). During the friction of the quenched sample, the microstresses in 7.5 and 0.5 txm-deep layers were less than in the layers more than 4 tzm in thickness in the brass-plated sample (the distortions of the crystal lattice Ad/d differ by about a factor of two). In the subsurface layers (thickness h -~ 1 /zm), inhomogeneities in the elemental composition and a stressed state [7] were observed which are typical for processes of contact interaction whereas there were no basic differences between the curves 3 in Fig. 6c since they could stem from the residual effect of the frictional coating. Thus, the aggregate of data on structural changes in the deformation zone of samples of the 50G steel subjected to quenching with rf heating and the combined treatment (RFH + FANT) and treated on a laboratory-scale friction-test machine indicates that there was no effect of the prior deposited film on the steady-state conditions of the friction. At present, while using commercial engines as well as while repairing them, linings are used that are quenched with rf heating. In spite of the great deformation of the cylinder face (compared to crude steel or cast iron), quenching stresses, and greater cost of manufacturing, quenched parts have high wear resistance. But on the operating cylinder-lining--piston-ring surfaces, the probability of the appearance of polishing defects is greater (specific surface defects), which can give rise to the generation of fins [12]. In order to eliminate these defects, finishing antifriction nonabrasive treatment has begun to be used recently [6]. In addition, an experiment on the application of the FANT technology [13] indicates that, upon proper selection

~(hkD

133

of the treatment conditions, the frictional forces, wear intensity, the tendency toward their seizure, oxidation processes on the surface, as well as the intensity of fatigue fracture decrease. The most substantial indication of indications of the positive role of FANT of quenched linings are presented in [14] in which, from an analysis of the properties of the microgeometry and surface-profile curves, it was shown that such a treatment has a favorable influence on the conditions of contact between a piston ring and a lining. In this connection, one may obviously expect that the greatest effect of FANT of quenched linings will be on the stage of breaking in under test conditions or during the operation of a real generator. Indeed, this has, above all, been confirmed in practice [4, 14]. In [13], the basic factors influencing the quality of the deposited layer have been analyzed and it has been noted that the nature of distribution of the coating on the surface of steel and cast-iron parts significantly depend on their composition. Our experimental data indicate the fundamentally different contributions of FANT to the frictional properties of quenched and normalized linings made of 50G steel under identical parameters of this treatment. It has been shown in [8] that, upon contact interaction between a normalized lining and a ring, a constant process of penetration of the elements takes place from the friction rod to deformed microvolume and secondary modified zones are formed. These zones have special characteristic structural parameters in conjunction with the increased wear resistance produced. The aftereffect is the most significant among the factors that control the longevity of the pair [8]. During the friction of a quenched steel with the coating (see Fig. lb), up to 75 h a gradual process of increase in the surface area of contact took place because of the involvement of regions with a soft layer and, hence, due to an inadequate "pliability" of the matrix, it fractured upon repeated deformation. The period of breaking in was related to this whereas, after 75 h of wear, the matrix (steel) was evidently the primary material that was subjected to wear. Therefore, the role of the coating consisted of a gradual formation of a surface area of the actual contact (failure occurred mainly on account of the coating) whereas, under steady-state conditions, the mechanism of contact was controlled by the matrix (wear intensity was practically the same as in the material without any coating). The results of investigation of the combined treatment (RFH + FANT) of the 50G steel indicated that there was no effective transition (modified) layer and, therefore, a large enough adhesive force between the film and the substrate which, as a rule, must increase in the presence of such a layer consisting of a compound based on the components of the matrix and the friction rod. The distinguishing features of the coating on the quenched steel were as follows. Low values of the function/~(hkl) ----fir) of the steel for all values of the argument compared to the steel without any coating; the ratio between the gradients of /3(~t) on the layers subjected to the action to its value after the treatment was close to unity; inversion of the break on the Hall dependences compared to those in the normalized state; a sharp decrease in the absolute variation in the nature of the crystal lattice of the steel upon FANT. These features were observed in the presence of the characteristic signs of modification of the surface microvolumes [8] but they were not very pronounced. It is logical to assume that it was determined by the properties of the material treated. The incorporation of atoms of the friction rod (and, possibly, also of elements of the technical-grade liquid), which is a necessary condition for creating a modified zone and an aftereffect of the coating during the tests, becomes difficult in the structure of the quenched steel to a greater degree than for the normalized steel. The negative properties of the surface layers subjected to the action of the treatment and accordingly the limitation of time of the positive role played by brassplating were related to the absence of the formation of a "stable" modified zone. According to the depletion of the service life of the coating, the effective life of a pair is controlled by the conditions of the lining--ring contact without any coating, where structural changes occur in the deformation zone of the steel that are typical for friction under surface-lubrication conditions. Thus, the changes in the structure of the surface layer of a lining made of the 50G steel stemming from FANT treatment are not retained till the onset of the steady-state frictional conditions. The effect of the coating is manifest only in the stage of breaking-in when wear takes place only on the deposited layer of the brass. Conclusions. 1. In laboratory tests on lining--ring pair at 30 MPa pressure and an average velocity of reciprocating sliding friction of 0.1 m/sec in M-10G2K lubricant, the 50G steel after RFH and a combined treatment (RFH + FANT) in the frictional conditions established has close values of the intensity of wear. Under breaking-in conditions, the wear of the brassplated sample is greater and is basically determined by the breakdown of the deposited coating. 2. An analysis of the aggregate of structural parameters of the quality of the finishing, antifriction, nonabrasive surface treatment/3(hk/) = fit), /~2COS20/~k2 = f(Ksin20/~k2"y2), gradt/3(hkt), Aa = f(t) showed substantial distinctive signs which are the causes for the limitation of the effectiveness of the tribological brass plating of the quenched 50G steel under certain conditions of the treatment (the positive role is noted only in the stage of breaking-in).

134

3. Structural criteria are suggested for the quality of the coatings revealed with the help of a special method of nondestructive testing of the material of thin surface layers, which can be used as a rapid method for estimating the parameters of the technological process of the surface treatment.

REFERENCES

.

.

3. 4.

.

.

.

8.

.

10. 11. 12. 13. 14.

A. A. Polyakov, "A procedure for laboratory investigations of the wear of materials of piston rings and cylinders with the 77MT-1 machine," in: Abs. Conf. on "Methods of Wear Testing" [in Russian], Akad. Nauk SSSR, Moscow (1962), pp. 225-228. D. N. Garkunov (ed.), in: Selective Transmission in Heavily Loaded Friction Joints [in Russian], Mashinostroenie, Moscow (1982), pp. 17-22. G. Pol'tser, V. Mfiller, G. Rhein_hold, and I. Lange, "Friction on brass-plated surfaces of steel- and cast-iron parts," in: Longevity of Rubbing Machine Parts [in Russian], Mashinostroenie, No. 2, 81-85 (1987). A. G. Andreeva, F. Kh. Burumkulov, V. I. Tolokonnikov, et al., "Finishing, antifriction, nonabrasive treatment as a means of increasing the life of machines and equipment," in: Longevity of Rubbing Machine Parts [in Russian], Mashinostroenie, No. 4, 34-59 (1990). L. M. Rybakova, L. I. Kuksenova, Yu. A. Nazarov, et al., "Structural changes and wear resistance of the 50G steel after finishing, antifriction, nonabrasive treatment of the surface," in: Abs. Repts. at the IVth All-Union Conf. on Thermal Microscopy [in Russian], Inst. of Mech. Eng., Russ. Acad. Sci., Moscow (1992). B. V. Namakonov, V. V. Kisel', and V. P. Lyalyakin, "Increasing the longevity of cylinder linings of internalcombustion engines by FANT," in: Longevity of Rubbing Machine Parts [in Russian], Mashinostroenie, No. 4, 139144 (1990). L. M. Rybakova and L. I. Kuksenova, Structure and Wear Resistance of Metals [in Russian], Mashinostroenie, Moscow (1982). L. M. Rybakova, L. I. Kuksenova, and Yu. A. Nazarov, "Structure of 50G steel after finishing, antifriction, nonabrasive treatment of the surface and estimation of the quality of the modified layer," Metalloved. Term. Obrab. Met., No. 10, 14-20 (1992). D. H. Buckley, "Adhesion of metals to a clean iron surface studied with LEED and Auger emission spectroscopy," Wear, 20, No. 1, 89-102 (1972). E. A. Marchenko, Failure of the Metal Surface during Friction [in Russian], Nauka, Moscow (1979). Ya. M. Golovchiner, "Procedure of determination of type-II stresses and the mosaic size," Zavod. Lab., No. 4, 431444 (1960). M. A. Grigor'ev, K. A. Verner, V. D. Zelenova, et al., "Abrasion-resistance of quenched cylinder linings and the criteria for its estimation," Vestn. Mashinostr., No. 11, 19-21 (1986). G. Pol'tser, A. Firkovskii, I. Lange, et al., "Finishing, antifriction, nonabrasive treatment (FANT) and selective transmission," in: Longevity of Rubbing Machine Parts [in Russian], Mashinostroenie, No. 5, 86-122 (1990). F. Kh. Burumkulov, A. G. Andreeva, A. T. Kulakov, et al., "Bench tests without using any motor of cylinderlining--piston-ring pair of the KAMAZ-740 engine," Vestn. Mashinostr., No. 3, 39-4I (1992).

135