Room temperature ionic liquids as lubricant additives in steel–aluminium contacts: Influence of sliding velocity, normal load and temperature

Please download to get full document.

View again

All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
 38
 
  Room temperature ionic liquids as lubricant additives in steel–aluminium contacts: Influence of sliding velocity, normal load and temperature
Share
Transcript
  Wear 261 (2006) 347–359 Room temperature ionic liquids as lubricant additives in steel–aluminiumcontacts: Influence of sliding velocity, normal load and temperature A.E. Jim´enez, M.D. Berm´udez ∗ , F.J. Carri´on, G. Mart´ınez-Nicol´as Grupo de Ciencia de Materiales e Ingenier´ıa Metal´ urgica, Departamento de Ingenier´ıa, de Materiales y Fabricaci´ on,Universidad Polit´ ecnica de Cartagena, Campus de la Muralla del Mar, C/Doctor Fleming s/n, 30202-Cartagena, Spain Received 27 May 2005; received in revised form 8 November 2005; accepted 18 November 2005Available online 18 January 2006 Abstract 1- n -Alkyl-3-methylimidazolium  X  − [  X  =PF 6 ;  n =6 (L-P106).  X  =BF 4 ;  n =2 (L102), 6 (L106), 8 (L108).  X  =CF 3 SO 3 ;  n =2 (L-T102).  X  =(4-CH 3 C 6 H 4 SO 3 );  n =2 (L-To102)] and 1-butyl-3-methylpyridinium bis(trifluoromethylsulfonyl)imide (L-PY104) have been studied as 1wt.% baseoil additives in variable conditions pin-on-disk tests for AISI 52100 steel-ASTM 2011 aluminium contacts. Friction coefficients and wear ratesincrease under increasing normal loads. Effective lubrication is obtained for a 0.15–0.20ms − 1 sliding velocity. Low friction and minimum wearrates are achieved for all additives at 25 ◦ C, 2.45N and 0.15ms − 1 . A transition to an abrasive wear mechanism is observed at 0.06ms − 1 , both atroom temperature and at 100 ◦ C. Friction coefficients for IL addtives are similar or lower than for neat ILs, while wear rates for 1wt.% ILs can beseveral orders of magnitude lower than those for neat ILs. The exception is the long alkyl chain L108 which always shows better lubricating abilityas pure lubricant, probably due to its lower miscibility with the base oil. Energy dispersive (EDS) and X-ray photoelectron (XPS) spectroscopiesshow surface interactions but, in contrast with neat ILs, no tribocorrosion processes are observed for the IL additives. Aluminium corrosion is onlyobserved after 100h immersion in water with 1wt.% L102.© 2005 Elsevier B.V. All rights reserved. Keywords:  Aluminium; Steel; Ionic liquids; Friction; Wear; Lubrication; Corrosion; Additives 1. Introduction Because of their unique combination of properties, the non-volatile, thermally stable, highly polar room-temperature ionicliquids (ILs) are being studied as high performance base lubri-cants [1–10]. Their miscibility both with water and organic solvents could also be exploited in their use as lubricating addi-tives [11–13].Recent studies have raised interest on the solution propertiesof ILs. It has been shown [14,15] that small amounts of water in fluorinatedILs,suchas1-butyl-3-methylimidazoliumhexafluo-rophosphate and tetrafluoroborate, have a dramatic effect on therate of diffusion of neutral and ionic species in these media. Itwas thus proposed that “wet” ILs may not be regarded as homo-geneous solvents but have hydrogen-bond interactions that canpossess in some cases polar and non-polar regions [16]. ∗ Corresponding author. Tel.: +34 968325958; fax: +34 968326445.  E-mail address:  mdolores.bermudez@upct.es (M.D. Berm´udez). 1,3-Dialkylimidazolium salts form supramolecules in solidstate through hydrogen bonds, which are maintained to a greatextent even in solution in solvents of low dielectric constant,evidencedthroughthedetectionofstructurespossessingmono-,di-and tricharged species. Based on these observations, imida-zolium ILs mixed with other molecules have been defined asnanostructured materials [16].The use of these advanced fluids could contribute to solvesome difficult tribological problems such as the lubrication of aluminium–steel contacts [7,8,11].In our previous study [11], we have shown that the lubri- cating performance of neat ILs on Al–steel lubrication dependon experimental conditions and is related to alkyl chain lengthand anion composition. In the case of highly reactive tetrafluo-roborate and hexafluorophosphate ILs, we have shown that theycause severe tribocorrosive attack when used as neat lubricantsof the steel–aluminium contact. These tribocorrosion processescan be detected by real-time observation of friction incrementsand wear debris precipitation. In that same study, when tworeactive ILs (L102, L-P106) were used as lubricant additives, 0043-1648/$ – see front matter © 2005 Elsevier B.V. All rights reserved.doi:10.1016/j.wear.2005.11.004  348  A.E. Jim´ enez et al. / Wear 261 (2006) 347–359 Table 1Ionic liquids used as 1wt.% additives in Al 2011-AISI 52100 steel lubricationImidazolium derivativesIL  R 1  R 2  X  − L102 CH 3  C 2 H 5  BF 4 − L106 CH 3  C 6 H 13  BF 4 − L108 CH 3  C 8 H 17  BF 4 − L-T102 CH 3  C 2 H 5  (CF 3 SO 3 ) − L-To102 CH 3  C 2 H 5  (4-CH 3 C 6 H 4 SO 3 ) − L-P106 CH 3  C 6 H 13  PF 6 − Pyridinium derivativeILL-PY104 tribocorrosion processes were not observed and a more effec-tivelubricationwasachievedwithlowerfrictionandwearvaluesthan those of the neat ILs.In the present work, we extend the study of ILs as lubricantadditivesinAl–steelcontacttoaseriesofimidazoliumandpyri-dinium salts under variable sliding velocity, normal load andtemperature in order to determine the conditions of effectivelubrication, the different wear mechanisms and surface interac-tions, and wear regime transitions. 2. Experimental details Ionic liquids (Table 1) were commercially available from Fluka Chemie (Germany). ASTM 2011 (5.62% Cu; 0.50% Fe;0.49% Bi; 0.49% Pb; 0.49 Si; 0.10% Zn) aluminium disks(44mm diameter; 10mm thickness) of hardness 107 HV weretested in a pin-on-disk tribometer (Microtest, Spain) againstAISI 52100 (1.52% Cr; 0.95% C; 0.33 Mn %; 0.25% Si) steelballs (0.8mm sphere radius) of hardness 848HV, in the pres-ence of 2ml of an additive free paraffinic-naftenic mineral baseoil [17,18] modified by the addition of a 1wt.% ratio of IL.Tribological tests according to ASTM G 99–04 standard werecarried out in air for a sliding distance of 850m. Variable slid-ing speed tests (0.06; 0.15; 0.20ms − 1 ) were performed at roomtemperatureunder2.45N(0.92GPa).Variablenormalloadtests Table 2Viscosity valuesIL (1wt.% in base oil) Viscosity (mPas) (standard deviation)25 ◦ C 100 ◦ CL102 324.7 (0.50) 22.1 (0.60)L106 313.3 (0.43) 19.8 (0.29)L108 311.8 (0.12) 18.7 (0.79)L-P106 309.8 (1.91) 18.7 (0.07)L-T102 303.6 (2.24) 15.6 (0.39)L-To102 296.3 (0.97) 14.6 (0.05)L-PY104 301.4 (1.02) 14.5 (0.32) (2.45; 3.45; 4.45N) were carried out at room temperature and0.15ms − 1 . For tests at 100 ◦ C, a 0.06ms − 1 sliding speed anda 2.45N normal load were selected. Friction coefficients werecontinuously recorded with sliding distance for each test. Vol-umelossfromaluminiumdiskswasdeterminedfromweartrack widthmeanvaluesafteratleast12measuresalongthewearscarswith standard deviation lower than 3%. Mean friction coeffi-cients and wear rates are obtained after at least three tests underthe same conditions.Viscosity values (Table 2) correspond to mean values after three measurements and were obtained in air for 12ml volumesamples using a HAAKE viscosimeter with maximum rotorspeed of 300s − 1 .ImmersioncorrosiontestshavebeencarriedoutaccordingtoASTM G 31-71 standard at room temperature under stirring at500rpm by means of a magnetic stirrer.Optical micrographs were obtained using a Leica DMRXoptical microscope. SEM images and EDS analysis wereobtainedusingaHitachiS3500Nscanningelectronmicroscope.The binding energies (Table 3) were obtained using a Physical ElectronicsSystemESCA5701spectrometerandarereferencedto the C 1s peak (284.6eV) used as internal standard. The pre-cision in the binding energy is estimated to be ± 0.1eV. 3. Results and discussion 3.1. Influence of sliding velocity Fig. 1 shows friction coefficients (Fig. 1a) and aluminium disks wear rates (Fig. 1b) variation with sliding speed for the seven 1wt.% IL additives.As shown in Fig. 1a, under low load (2.45N; 0.92GPa) the highest friction values are obtained at the lowest speed(0.06ms − 1 ), which also corresponds to maximum wear ratesfor all additives. Table 3XPS binding energies (eV) for the wear debris (4.45N; 0.15ms − 1 ; 25 ◦ C)IL additive (1wt.%) Al (2p) Cu (2p) F (1s) O (1s) N (1s) Fe (2p)L102 75.6 (33.20%) 934.1 (0.98%) 687.0 (27.79%) 533.0, 531.9 (34.21%) 401.4 (3.68%) 711.8 (0.19%)L106 75.3, 72.9 (22.88%) 933.6 (0.55%) 686.7 (7.21%) 532.1, 530.1 (61.44%) – 710.6 (2.04%)   A.E. Jim´ enez et al. / Wear 261 (2006) 347–359  349Fig. 1. (a) Friction and (b) wear rates as a function of sliding speed (2.45N;25 ◦ C). Table4comparesfrictionandwearvaluesforneatand1wt.%ILs as a function of sliding speed.As can be seen on Table 4, in general, all ILs additives give similar or lower friction values and up to three orders of magni-tude (in the case of L102, at 0.15ms − 1 ) lower wear rates thanneat ILs.The lowest friction in the whole range of velocities isobserved for two reactive ILs, the short alkyl chain tetrafluo-roborate derivative L102 and the hexafluorophosphate L-P106additives, which were the first ILs studied as lubricant oil addi-tives [11].Aswehaverecentlydescribed[11]andcanbeseeninTable4, this good lubricating ability of L102 and L-P106 additives is incontrast with the high friction and wear obtained for these twoILs when used as neat lubricants due to the severe tribocorro-sion processes which take place at the Al–IL–steel interface.The results presented here support the hypothesis that a 1wt.%proportion of IL is enough to form an ordered adsorbed bound-ary layer without producing severe tribocorrosion processes.At low speed (0.06ms − 1 ), the lowest wear rates are obtainedfor the hexyl-substituted imidazolium derivatives 1wt.% L-P106 and L106; the order from lower to higher wear rates forthe seven additives being: L-P106<L106<L102<L-T102<L-To102<L-PY104<L-108 (see Fig. 1b). This is in contrast with  T   a     b     l   e     4     I   n     fl   u   e   n   c   e   o     f   s     l     i     d     i   n   g   v   e     l   o   c     i    t   y   o   n    t     h   e     f   r     i   c    t     i   o   n   a   n     d   w   e   a   r   o     f     A     l  –   s    t   e   e     l     f   o   r   n   e   a    t     (     1     0     0     %     )   a   n     d     1   w    t .     %     I     L   a     d     d     i    t     i   v   e   s     N   o   r   m   a     l     l   o   a     d   =     0 .     9     2     G     P   a   ;    t   e   m   p   e   r   a    t   u   r   e   =     2     5     ◦      C     I     L     0 .     0     6   m   s    −      1      0 .     1     5   m   s    −      1      0 .     2     0   m   s    −      1     µ      K      (   m   m      3    m    −      1      )     µ      K      (   m   m      3    m    −      1      )     µ      K      (   m   m      3    m    −      1      )     1     0     0     %     1     %     1     0     0     %     1     %     1     0     0     %     1     %     1     0     0     %     1     %     1     0     0     %     1     %     1     0     0     %     1     %     L     1     0     2     0 .     2     0     9     0 .     0     7     8     1 .     6     5     ×      1     0    −      3      8 .     5     1     ×      1     0    −      5      0 .     2     8     4     0 .     0     7     0     6 .     6     2     ×      1     0    −      3      6 .     7     6     ×      1     0    −      6      0 .     2     6     0     0 .     0     6     6     5 .     4     5     ×      1     0    −      3      1 .     8     1     ×      1     0    −      5      L     1     0     6     0 .     0     7     1     0 .     1     3     2     1 .     5     4     ×      1     0    −      5      4 .     3     2     ×      1     0    −      5      0 .     1     1     5     0 .     1     1     0     1 .     5     4     ×      1     0    −      5      1 .     4     0     ×      1     0    −      5      0 .     0     6     1     0 .     1     1     7     6 .     8     2     ×      1     0    −      5      1 .     6     9     ×      1     0    −      5      L     1     0     8     0 .     0     4     4     0 .     1     1     2     7 .     5     7     ×      1     0    −      6      1 .     9     9     ×      1     0    −      4      0 .     0     2     1     0 .     0     7     9     1 .     0     9     ×      1     0    −      5      2 .     2     9     ×      1     0    −      5      0 .     0     9     6     0 .     1     0     7     8 .     4     0     ×      1     0    −      6      3 .     7     1     ×      1     0    −      5      L  -     T     1     0     2     0 .     1     1     8     0 .     1     0     0     1 .     2     1     ×      1     0    −      3      1 .     3     0     ×      1     0    −      4      0 .     1     0     3     0 .     1     0     1     1 .     6     9     ×      1     0    −      3      1 .     1     6     ×      1     0    −      5      0 .     0     8     9     0 .     0     8     5     0 .     9     5     ×      1     0    −      3      1 .     6     7     ×      1     0    −      5      L  -     T   o     1     0     2     0 .     1     3     0     0 .     1     2     5     0 .     4     9     ×      1     0    −      3      1 .     6     6     ×      1     0    −      4      0 .     1     0     1     0 .     0     9     7     7 .     7     3     ×      1     0    −      4      1 .     9     3     ×      1     0    −      5      0 .     1     3     8     0 .     1     1     0     0 .     9     2     ×      1     0    −      3      1 .     6     1     ×      1     0    −      5      L  -     P     1     0     6     0 .     1     0     7     0 .     1     0     2     2 .     2     2     ×      1     0    −      3      2 .     8     3     ×      1     0    −      5      0 .     1     6     3     0 .     0     6     6     1 .     4     7     ×      1     0    −      3      1 .     6     5     ×      1     0    −      5      0 .     0     6     2     0 .     0     5     8     1 .     2     0     ×      1     0    −      3      3 .     5     7     ×      1     0    −      5      L  -     P     Y     1     0     4     0 .     1     1     3     0 .     1     3     1     7 .     6     4     ×      1     0    −      4      1 .     7     6     ×      1     0    −      4      0 .     2     1     7     0 .     1     1     3     2 .     8     3     ×      1     0    −      3      1 .     4     9     ×      1     0    −      5      0 .     1     7     6     0 .     1     1     1     1 .     7     5     ×      1     0    −      3      1 .     6     9     ×      1     0    −      5  350  A.E. Jim´ enez et al. / Wear 261 (2006) 347–359 the order obtained for neat ILs under the same conditions [11](see Table 4).The long chain imidazolium derivative L108 shows the bestlubricatingabilityinthepurestate(seeTable4).Thisisinagree- ment with previous observations [7,11] which have shown thationic liquids of longer alkyl chain have stronger adsorption onthe sliding metallic surfaces and chemisorbed films are easierto form and hence they show better antiwear ability than theones with shorter alkyl chain. However, L108 shows a poor per-formance when used as lubricant additive, particularly at lowvelocity and room temperature (see Table 4 and Fig. 1a and b). This could be attributted to its lower miscibility with the baseoil, which prevents the formation of an effectively lubricatingadsorbed layer.In general, imidazolium 1wt.% ILs give lower friction andwear rates than the pyridinium salt L-PY104. This could bedue to the higher hydrophobicity of the bis(trifluoromethyl-sulfonyl)imide derivative L-PY104 compared to that of thetetrafluoroborateILs[14],asthetestsarecarriedoutinair.Ithas been recently shown [14] that the viscosity of hydrophobic ILs decrease in the presence of water. As can be seen on Table 2,the viscosity values for the base oil with 1wt.% L-PY104 isamong the lowest of the series of IL-modified lubricants studiedhere.When sliding velocity is increased to 0.15ms − 1 , a minimum(10 − 6 –10 − 5 mm 3 m − 1 ) mild wear (see Table 4 and Fig. 1b) value is obtained for all ionic liquid additives independent of composition. A slight increase is observed when the speedis increased at 0.20ms − 1 , although the mild wear regime ismaintained.This critical influence of sliding velocity on wear rates is notobserved in the case of the neat ILs [11] (see Table 4). Interestingly, in a previous work  [18] with several liq-uid crystalline additives, the ionic liquid crystal additive  n -dodecylammonium chloride showed this same wear variationwith sliding speed, under the same set of experimental condi-tions, with minimum wear rate at 0.15ms − 1 . This wear depen-dencewithvelocityisnotobservedwhenneatbaseoilorneutralliquid crystal additives are used [17].Thiscriticalvelocitywouldcorrespondtoconditionsofgoodmiscibility with the base oil and formation of the chemisorbedlubricating film for all additives on the metal. 3.2. Influence of normal load  While under low or moderate loads, an effective adsorbedboundarylubricatingfilmcouldform,ahighpressurecouldgiverise to tribochemical and corrosive effects.The0.15ms − 1 criticalvelocityforlowfrictionandminimumwear was selected to carry out variable load tests.Both friction (Fig. 2a) and wear (Fig. 2b) increase under increasing loads. Under 4.45N, a maximum friction coef-ficient value of 0.131–0.138 and a severe wear regime(1.4–2.0 × 10 − 3 mm 3 m − 1 ) are obtained for all IL additives.Thus, under this critical pressure the disruption of the adsorbedlubricating films takes place, regardless of composition. Thehighest wear rates at 4.45N (Fig. 2b) correspond to 1wt.% L- Fig. 2. (a) Friction and (b) wear rates as a function of normal load (0.15ms − 1 ;25 ◦ C). To102andL-PY104additives,preciselythosewhichgivelowerviscosity values (Table 2). Those ILs with fluoride or phospho- rus containing reactive anions (L-P106; L-T102) and/or longeralkyl chains (L108) show better load-carrying capacity. 3.3. Influence of temperature Underlowload(2.45N)andlowslidingvelocity(0.06ms − 1 ),increasing test temperature to 100 ◦ C gives friction values(Table 5) similar or lower to those obtained at room temper- ature (Table 4). Notable exceptions are the best lubricating Table 5Friction and wear of Al–steel for 1wt.% IL additives at 100 ◦ C (0.06ms − 1 ;2.45N) (standard deviations in parenthesis)Additive (1wt.%)  µ  K   (mm 3 m − 1 )L102 0.113 (9.24 × 10 − 3 ) 2.40 × 10 − 4 (4.76 × 10 − 5 )L106 0.122 (7.53 × 10 − 3 ) 1.57 × 10 − 4 (1.55 × 10 − 5 )L108 0.079 (7.12 × 10 − 3 ) 1.60 × 10 − 4 (1.40 × 10 − 5 )L-T102 0.083 (2.48 × 10 − 3 ) 1.64 × 10 − 4 (2.70 × 10 − 5 )L-To102 0.083 (9.65 × 10 − 3 ) 1.46 × 10 − 4 (1.55 × 10 − 5 )L-P106 0.146 (1.51 × 10 − 2 ) 1.41 × 10 − 4 (7.45 × 10 − 6 )L-PY104 0.078 (4.78 × 10 − 3 ) 1.50 × 10 − 4 (2.46 × 10 − 5 )   A.E. Jim´ enez et al. / Wear 261 (2006) 347–359  351Fig. 3. Wear tracks and steel balls (2.45N; 0.06ms − 1 ; 25 ◦ C): (a) 1% L-P106; (b) 1% L106; (c) 1% L108; (d) 1% L-To102; (e) 1% L-PY104.
Related Search
Similar documents
View more
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks
SAVE OUR EARTH

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!

x