Downloaded by [75.149.200.233] at 08:27 25 July 2012

Effects of Grease Composition and Structure on Film Thickness in Rolling contacta I. COURON* and P. VERGNE LMC, I.E.T., UMR CNRS/INSA 55 14 MSA Bit. 113 69621 Villeurbanne cedex France and D. MAZUYER LTDS, l.E.T., UMR CNRSIECL 5513, ECL 69 131 Ecully cedex France and N. TRUONG-DINH CONDAT, BP 16 38670 ChasselRhBne France and D. GIRODIN SNR Roulements, BP 2017 74010 Annecy cedex France

The aim of this work is to show the correlation between the tribological behavior of grease and its composition and structure. A tribological investigation was conducted on various lubricants. The following parameters were varied: base oil, soap and presence of additives. To etlsrrre eficient control of grease composition, greases containing the same type of soap were manufactrcred from the sanle concentrated soap sample. Film thickness measurements showed that the thickener nlicrostructure (revealed by TEM observations) is not the determining factor for thefornlation of a thick lubricantfilm, i.e. a film follondng EHL equations. Nevertheless, the soap - base oil interaction is an essential parameter. The composition of a grease influences oil bleeding, mechanical stability, and rheological behavior. The elastic n~odulusG' seems to be the only parameter directly linked to tribological behavior. Greases with low G' have a greater capacity to fornz a thick EHDfilm compared to greases nith large G'.

Final manuscript approved June 7,2002 Review led by Paul Bessette

KEY WORDS Grease Thickness

Lubrication;

Microstructure;

Rheology;

Film

IN'TRODUCTION Grease is a complex multi-phase material whose way of functioning needs to be clarified because of its growing use in modem machines. Presently, the tribological behavior of grease in a lubricated contact is not completely understood ( I ) . Nevertheless, people in industry expect to predict final grease behavior in a contact from knowledge of its initial composition. Some authors show the influence of experimental conditions on film thickness ( l l ) ,(6). Hurley and Cann observed that this thickness increases when the rolling speed increases. Others studied the influence of composition (2), (10). Mitsui and Spikes noticed that polymer macromolecules present in an oil can be adsorbed onto the surfaces and thus promote film formation. All these previous works consider the grease as a single entity. No account is taken of the multi-phase aspect of this lubricant and the interaction between its components. The objectives of the authors' study are: Firstly, to analyze the influence of the interactions between grease components on the film thickness in a rolling contact. Secondly, to try to find how the ability of a grease to give fully flooded lubrication is related to its intrinsic properties.

I. COURONN~, P. VERGNE, D. MAZUYER, N. TRUONG-DINH AND D. GIRODIN

Downloaded by [75.149.200.233] at 08:27 25 July 2012

GREASE NAME LiMVT LiSV AlMV A l SVT

1

SOAP CONCENTRATION: BASEOIL VISCOSITY: 1 12% wlw 65cSr~rlO.C Lithium Lithium Ester Aluminum complex ( Mineral 11 Aluminum complex Ester

Glass plnle

-+

V:

VISCOSITY IMPROVER

ADHERENCE-~IMPROVER ~

V+T

V V V+T

The results of only four greases will be described in this paper: these lubricants have been chosen because they represent all the possible combinations between base oils and soaps. Their composition is given in Table 1.

Semi-reflecting mirror Microscope

1

lj

Methods

Lod system

DC mtor

Cr layer

Film Thickness Measurements Film thickness in the center of the contact (h,) was measured by optical interferometry (Fig. 1). A 52100 steel ball (25.4 mm diameter, mirror-polished with roughness s 20 nm) is loaded against a glass disk over which is installed a microscope to observe the contact zone. The contact is illuminated by white-light with normal incidence. The plate surface (60 mm diameter, roughness < 5 nm) in contact with the ball is coated with a thin chromium semi-reflecting layer (==10 nm thickness) to improve light reflection and thus the contrast of the observed image. A color CCD video-camera is mounted on the microscope eyepiece and is connected to a video tape recorder and to a monitor. This system allows the visualization and recording of the contact zone throughout the experiment. The load imposed on the contact is low 15 N) and corresponds to a maximum pressure of 300 m a . The grease to be tested was spread over the whole surface of the plate. Its initial thickness was less than 1 mm and grease was not added during the tests. Experiments were done in pure rolling at 20°C. Computer software allows the fringes observed in the contact to be converted to film thickness. Film thickness was analyzed for speeds increased by stages of five minutes. The mean rolling speed (U,) was varied in the range 0.01 to 0.6 m/s.

0 DC motor

Fig. 1--Optical Interferometry apparatus.

SAMPLES AND METHODS Samples Usually, grease consistency is adjusted during manufacturing to reach a specified NLGl grade. Here, concentrations of the grease components were imposed and the NLGI grade was a consequence of this choice. In order to limit scattering due to grease manufacturing, greases containing the same type of soap were formulated from the same concentrated soap batch. The manufacturing procedure was the same for all the lubricants and included homogenization at the end of the process. The parameters varied were the base oil type (mineral (M) or ester (S)), the soap type (lithium (Li) or aluminum complex (Al)) and the presence of two tnacromolecular additives (V and T, respectively viscosity and adherence improvers). Mineral and ester base oil viscosity (65cSt at 40°C) and soap concentration (12% w/w) were the same for all the samples. When present, V and T concentrations were respectively 5.5 and 0.03 % w/w. Greases contain the same package of performance additives (anti-oxidant, anti-wear, extreme pressure,. ..). So, these greases are not model lubricants: they have a complete formulation allowing them to be used in rolling bearings.

(p.

Microsrructure and Physicochemical Properties of Greases Fresh samples were observed by TEM (Transmission Electron Microscopy) as follows: a small amount of grease was spread on a glass plate, then a TEM grid (of copper covered by a continuous carbon film) was delicately pressed onto the grease. The sample should not be sheared too much in order to protect the microstructure. The grid was then immersed in hexane for fifteen minutes to dissolve most of the base oil. The solvent was finally evaporated in an oven at 30 to 40°C for ten minutes and the covered grid was ready for TEM observations. Rheological measurements were conducted at 25OC under the controlled stress mode. Plate/plate tools were chosen with an

'I'>~~!,I 2.

I:I.,~.,-I

C ~ ~ ~ I ~ T ~ ~ I I I ~ c.\I,.>~'II> , X ~ ~ \ c t t ( ; K I \VI.

FCMVh I.,KWIC.IIIV

I:II >!

c.\$,$
:, Iuhric:in!

lilt11 i\ rcI:t!ccl 10

111~-curve

11, =

l'lL'-,j.

Ifrhc 11 = l'it!,,,i .,.

curvc ;Icrces with 1 cliffrrent. TIIC fil111!hickr~c~\s tncrc:lces wilh \ p c < l ;and rc:$cllcr :! rel:tlivcly lligh th~uhnc\*i h ~2 250 111111. In ;yreenlrnt !\ill1 tlic prcdtclinn, nf171. il incrcnrcq \vilh s slope ol(l.h7. choracteri.lic o f a l'olly I1o1KlellL'.ILO co~lt:icl.

E ~ h l .tc repons IIIC rcst~ltsoh~:~incd ;~iw 011 h l c ~ d ~~IICI ~ ~ 1x11. y 1 1 i I rollr rrpre~cn1;lli~cgrl';t*cs. 1nili:tIly. oil l>lccccnl\ hC :Lgoq>dir~ f l h cc:tp;tcil! t,f @rcx\c 1 0 I Oit% ~ IP ~h~i~p : l nh l : ;tnd ~ ~\o 11, luhr!c:#ic the ci?nl:sl. I

Kcfcrencc I.
AVII

I~*~.~,AT RI.lymer;alditive\. These h;wc hccn proposed on ihc h:t\ir ( r l cheniical mslysis of hleetl p n ~ l u c t sand therm:all? aged gre:l\e\ dcwrihed in (.iJ.

~I imporhtnt I

Downloaded by [75.149.200.233] at 08:27 25 July 2012

Efkcts nf Grease Composition and Slruclure on Film 'lhickness in R n l l i ~ ~ Con~:lr~ s

Cbmpition

75

Sarpbosc oil inlcracllon

Oil bleeding

lo form r itlllg tlmird film

la fwm o fully f l d c d film

Fig. bOp1lcaI micmscopy Images ot the reslduel film o l grease alter its passage in the EHD rlg.

1.i-h (... Al-S...

LiS

... AI-M

...

G' yy 9WO Pa

C'-9000P~

Fig. 3delallon batween phySlCOchemlstry and rheological and ldbologlcai behaviors.

dn,plrtr in ihc Henzian ~clnc.Rut in l n ~ h cncer. ~ h c\trtlr.lure ,,I' rhc lilal, cspcci:~ll!. on lhc \ides 1~1'1hc 1r;lrk. ~lepcnd*nn lhc stnlcrurc ol'greasc \vllicll i r rcl:lictl l o lllr n;!lurc 01 llle rn;fip (cnlloi