ESTIMATION MODIFICATION UNDER

OF A C T I V A T I O N

ENERGY

OF F R I C T I O N A L

BOUNDARY

LUBRICATION

R. M . M a t v e e v s k i i , V. L. L a s h k h i , and

OF C H E M I C A L

SURFACES CONDITIONS

I. A. Buyanovskii, A . B. V i p p e r

UDC 620.178.162:621.892

The antifriction and antiwear or extreme-pressure properties of oils containing a c t i v e additives depend to a considerable extent on c h e m i c a I activation of the additives. The use of additives in oiis leads to fl, e d e v e l o p m e n t of a c h e m i c a l reaction in the friction zone and a m o d i f i c a t i o n of the surface layers of the m e t a l . The a b i l i t y of additives to enter into reaction with a m e t a I depends on a number of energy factors, among which we should note the strength of the bond in the a d d i t i v e m o l e c u l e that, in the process of reaction with the metal, creates modified layers in the form of con.pounds of sulfur, phosphorus, or chlorine. Determination of the r e a c t i v i t y of c h e m i c a l compounds with respect to m e t a l as for example, the r e a c t i v i t y of lube oil additives, is an independent and rather complex field of investigation. In this connection it has been of interest to study the reactivity of hthe oil additives, from the starting point of direct t r i b o c h e m i c a l m e a s u r e m e n t s , by analogy with the determination of heat of adsorption [1]. 22~e simplest and most convenient approach to ~ i s problem proved to be the widely known method for estimating the temperature stability of boundary lubricant films [2]. The tests were performed in a K T - 2 tester with a four-roller scheme [3] providing point contact of the specimens. The specific load in the contact zone was I50 k g f / m m z, and the r e l a t i v e rate of d i s p l a c e m e n t of the specimens was 0.3 m m / s e c . The tests were run with stepwise increases in the temperature of the assembly over a rang e from 20 to 300~ The test period at each t e m p e r ature was 60 sec. In the interest of a c o m p l e t e and d e t a i l e d accounting for various factors, the c h e m i c a l c o m p o sitions of the robbing pairs and of the lubricant m e d i u m were varied substantially from e x p e r i m e n t to experiment. In particular, rubbing elements were fabricated from s~andard ShKh-15 [52100] steel and also from steels c o n t a i n ing from I to 10% chromium or nickel. The specimens after h e a t treatment had hardnesses between 500 and 900 k g f / m m 2 (HV3). The working surfaces of the specimens were ground and polished to a Class 12 finish. The c h e m i c a l l y a c t i v e compounds used in this work were a chlorine-containing additive (Additive XII, a c h l o r i n a t e d wax) and a sulfur-containing additive (LZ-23k); these additives were blended at respective concentrations of 1.5% and 1% in a p h a r m a c e u t i c a l white minera] oil (usa = 27.8 cSt). The c r i t i c a l temperature (Tcr) was registered for each test, this being the temperature corresponding to the start of a sharp change in the coefficient of friction (f); also, the wear of the rubbing specimens was measured after the end of each c y c l e of the tests. A t y p i c a l curve for the temperature d e p e n d e n c e of f, obtained by the method just described, is shown in Fig. 1. The process described by the section of the curve AB has been studied rather thoroughly [3]; in contrast, the process described by the section BC, including the t e m p e r a t u r e of c h e m i c a l reaction Tch, still needs a more d e tailed examination. This process is t y p i c a l l y observed with additives that are high in c h e m i c a l a c t i v i t y ; it is r e lated to irreversible processes taking p l a c e on the frictional surface [4]. We have a t t e m p t e d a new approach to analysis of the results, starting from the following ideas. It is known that Ter corresponds to a degree of coverage of the rubbing surfaces equal to 0.5 [5], or, according to more recent data, as l i t t l e as 0.01-0.05 [6]. Evidently the increase in f in the region of temperatures above the c r i t i c a l (Fig. 1, section AB) is explained by subsequent d e sorption of additive from the frictional surface. Here the rate of c h e m i c a l reaction is too low to be of p r a c t i c a l importance. Beginning at point B, however, the d e v e l o p m e n t of c h e m i c a l processes becomes appreciable, and this

Translated from Khimiya i Tekhnologiya Topliv i Masel, No. 2, pp. 50-52, February, 1926.

9 1976 Plenum Publishing Corporation, 227 West 1 7th Street, New York, N. Y. 10011. No part o f this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, microfilming, recording or otherwise, without written permission o f the publisher. A copy o f this article is available from the publisher for $15.00.

151

0,35

1,5

o25 o,z8 8J5

Y

l

B

~ta ,

I

I

L

Hi content, at.

o,3a~

os' l J

I I

t --.

60

t5-~ i° <

020 l

OJ5°t g z, 8 8 I0

[

Cr content, at. % Fig. 1

Fig. ~

Fig. 1. Typical curve for temperature dependence of coefficient of friction. Fig. 2. Relation between activation energy of additive/metal interaction process and amount of wear: a) for system consisting of pharmaceutical white oil with 1°7oadditive LZ-23k and nickel-alloy steel; b) for system consisting of pharmaceutical white oil with 1.5% Additive XlI and chrome-alloy s t e e l leads to a perceptible decrease in f (Fig 1, section BC). Iri this region, the chemical reaction rate probably increases in proportion to the increase in temperature, this reaction leading to more and more modification of the frictional surface and a drop in f. At the POint C, which corresponds to Teh , a certain equilibrium has apparently been established between the rate of c h e m i c a l reaction (Vch) and the rate of wear (Vw). Consequently, we may assume that, in the section BC, the reaction rate Vch is inversely proportional to the change in frictional coefficient Af; i.e., Vcb ~c 1/Af. The specific rate of any chemical reaction can be represented by the equation [7]

V = ke-Zact/Rr where k is the preexponential factor; Eac t is the activation energy; R is the gas constant; T is the temperature. When this relation is plotted on coordinates of in V vs l / T , we can use the slope to determine the value of Eac t. By means of an analogous plot for the coordinate system log (1/&f) vs 1 / % we can obtain the activation energy for the process of chemical modification of frictional surfaces by additive molecules. This approach has been used in the present case in calculating the value of Eac t for systems consisting of chrome-alloy steel and Additive XlI. The values of Eac t in relative units and the amounts of wear for these systems are shown as functions of the content of alloying element in Figs. 2a and 2b. It will be seen from Fig. 2 that an increase in activation energy of the chemical interaction process is a c c o m panied by a decrease in wear, and conversely, a decrease in activation energy is accompanied by an increase in wear. This relationship is entirely logical if we consider that, in the presence of highlyreactive compounds, the wear that is observed in mainly chemical. Chemical modification of a frictional surface, as is well known, is preceded by adsorption of the particular compound at the interface and by the overcoming of the energy barrier that arises, i.e., by transition of the adsorbate to such a state that all the conditions necessary for chemical reaction are established. For monotypical additives with essentially identical mechanisms of action, the activation energy for the process of chemical interaction can evidently be correlated with the strength of the bond between the active element of the additive and the organic radical. The weaker this bond, therefore, the more intense should be the chemical reaction (with other conditions equal). This will be manifested most clearly in cases in which the steric factor (nature of molecular orientation in the boundary film) is essentially identical for the compounds being compared, so that no substantial changes are introduced into the values of Eac t or the preexponential factor.

152

With the aim of confirming this view, we selected two organosulfur compounds (diphenyl disulfide and dibenzyl disulfide) having essentially the same molecular diameter (6.6 and 6.4 A, respectively) but differing in R-S bond strength [8]. The bond strength (E b) was calculated, starting from a corollary of Hess' law; the vaiue for diphony1 sidulfide E~ = 108 keal/mole, and for dibenzyl disulfide E~ = 95 keal/mole. These compounds were blended at a concentration of 1% in pharmaceutical white oil. The tests were performed in a KT-2 four-ball friction tester [9] with standard balls made of ShKh-15 [52100] steel, 8 mm in diameter; the temperature of the friction assembly was increased in steps from 20 to 300~ the axial load was 11 kgf (average initial pressure at the contact 210 kgf/mm2), and the sliding speed of the upper ball was 0.04 cm/sec. The test period at each temperature was 60 sec, the same as in the preceding experiments. The results from these tests indicated that Tcr for the oil containing diphenyl disulfide was 130~ that for dibenzyl disulfide only 80~ However, the chemical activity of the oil with dibenzyl disulfide was somewhat higher; this can apparently be explained by the lower vatue for the m a x i m u m coefficient of friction (0.28 vs 0.34 for the diphenyl disulfide) and the greater rate of chemical reaction of the additive with the metal. The activation energy of the chemical interaction process in relative units, as caIeulated by the scheme we have proposed here, is E'~ct = 55 for the dibenzyl disulfide. Thus, on the basis of equality of the ratios Eb/E~b Eact/E" act' we can conclude that the above assumptions are valid. The higher value of Tcr for the diphenyl disulfide is apparently related to its greater adsorptive tendencies. Comparing the values for heat of adsorption(Qads) of the test compounds on FezO3 [10] (5.4 x 10 "3 c a l / g for diphenyl disulfide and 1.4 x 10 "a c a l / g for dibenzyl disulfide) with the values for critical temperature that we have given previously, we obtain the relationship r

r

T_.Cr

InQads

Ter

lnQads

which is in agreement with the theoretical premises for the work reported in [5, 6, 11]. In conclusion, we should note that such an approach to the prediction of service properties for lube oil additives is most convenient when studying compounds that are similar in composition and structure; the approach opens up new possibilities in using the temperature method to Study boundary lubricant films. LITERATURE

CITED

i. 2.

]. J. Frewing, Proc. Roy. Soc. A, 18___22,270 (1944). R.M. Matveevskii, Temperature Stability of Boundary Lubricant Films and Solid-Film Lubricant Coatings in Friction of Metals and Alloys [in Russian], Nauka, Moscow (1971).

3. 4. 5. 6. 7. 8. 9.

I . A . Buyanovskii, V. I. Kashin, L. B. Kraposhina, e t a l . , Mashinovedenie, No. !, 83 (1972). R.M. Matveevskii, Yu. A. Lozovoi, E. S Shepeleva, e t a l . , Khim. Tekhnol. Top1. Masel, No. 8, 39 (1970). E.P. Kingsbury, ASLE Trans., 3, No. 1, 30 (1960). L S. Akin, ASME Trans. B, 9_~5,No. 4 (1973). S. Gtasstone, K. L Laidler, and H. Eyring, The Theory of Rate Processes, McGraw-HilI, New York (1941). E.S. Forbes and A. L D. Reid, ASLE Trans., 16_, No. 1, 50 (1973). R.M. Matveevskii, KT-2 Four-Ball Friction Tester for Determining Critical Temperature of Oil Film on Metal [in Russian], VINITI, Moscow (1957). K . G . Allure and E. S. Forbes, J. Inst. Petr., 53._~,No. 521, 173 (1967). C . N . Rowe, ASLE Trans., 13_, No. 3 (1970).

10. 11.

1,53