Wear 292–293 (2012) 25–32

Contents lists available at SciVerse ScienceDirect

Wear

0043-16

http://d

n Corr

E-m

journal homepage: www.elsevier.com/locate/wear

Modification of polyetherimide by phenylethynyl terminated agent forimproved tribological, macro- and micro-mechanical properties

Ting Huang, Pei Liu, Renguo Lu, Zhongyuan Huang, Haiming Chen, Tongsheng Li n

State Key Laboratory of Molecular Engineering of Polymers, Department of Macromolecular Science, Fudan University, Shanghai 200433, China

a r t i c l e i n f o

Article history:

Received 14 November 2011

Received in revised form

31 May 2012

Accepted 1 June 2012Available online 13 June 2012

Keywords:

Sliding wear

Polymers

Wear testing

Nanoindentation

48/$ - see front matter & 2012 Elsevier B.V. A

x.doi.org/10.1016/j.wear.2012.06.003

esponding author. Fax: þ86 21 51630401.

ail address: lits@fudan.edu.cn (T. Li).

a b s t r a c t

The incorporation of 4-phenylethynylphthalic anhydride (4-PEPA) was utilized to synthesize a series of

phenylethynyl terminated oligomers, according to different calculated number-average molecular

weight (Mn). Modified polyetherimide (PEI) resins were prepared by the thermal curing reaction of

phenylethynyl terminated etherimide oligomers. For comparison, unmodified PEI was also prepared.

Thermal, macro- and micro-mechanical characterizations showed that the incorporation of phenyl-

ethynyl terminated agent resulted in a significant increase in glass transition temperature (Tg), tensile

strength and microhardness. The tribological properties of PEI specimens were evaluated on a

reciprocating friction and wear testing machine. Higher wear resistance was achieved by the

incorporation of phenylethynyl terminated agent. Furthermore, wear rate of modified PEI prepared

from the oligomers with the calculated Mn of 5000 g/mol was about five times lower than that of

unmodified PEI. At least for the range of this study, while increasing calculated Mn of oligomers, wear

rate of modified PEI initially decreased sharply and then increased slightly. This is clearly related to the

change in the macro- and micro-mechanical properties. Scanning electron microscope (SEM) observa-

tions of worn surfaces showed that type of wear changed from abrasive wear of unmodified PEI into

fatigue wear.

& 2012 Elsevier B.V. All rights reserved.

1. Introduction

Aromatic polyimide (PI) resins exhibit a number of outstand-ing physical and chemical properties, such as excellent thermaland thermo-oxidative stability, solvent resistance, as well asmechanical, anti-wear and electrical properties. These excellentproperties widely broaden the applications in automotive, bear-ings, micro-electronic appliances, aerospace materials and rail-way transport systems [1–3]. According to the increase in thedemand for the industrial applicability, significant efforts havebeen focused on preparing PI resins with superior mechanical andtribological properties.

In previous literatures, many researchers have investigatedvarious factors affecting tribological behaviors of PI, includingtype of atmosphere, temperature and contact stress [4–6].Furthermore, the addition of traditional micro-fillers, such asgraphite, MoS2 and fiber, could enhance tribological and mechan-ical properties [7–10]. Samyn and Schoukens [8] have proposedsome tribological mechanisms of thermoplastic PI modifiedPTFE. What is more, a great variety of nano-particles, such asmetal oxidation nanoparticles, carbon nanotubes (CNTs), silicon

ll rights reserved.

compounds and some other nano-particles could largely improvethe tribological properties of PI as well as mechanical properties[11–17]. Sinha and coworkers [16] have investigated the influenceof single walled carbon nanotubes (SWCNTs) addition on thetribological properties of the PI films and found that the incorpora-tion of the SWCNTs obviously enhanced the load-bearing capacityof the PI composites. On the other hand, only a few reports,however, have been available on the improvement of tribologicalbehaviors of PI modified by changing macromolecule structures.

Chitsaz-Zadeh and Eiss [18] have changed the diamine groupsinto structures with different elasticity modulus and found thatflexible chain configurations with a high toughness led to apositive effect on the wear rates of PI films. In our team group,Tian et al. [19] investigated the tribological properties of fluori-nated PI. Experimental results indicated that the friction coeffi-cient and wear rate decreased with increasing fluoride monomerratio. These above researches have indicated that mechanical andtribological properties were depended heavily on the structures of PI.Thereby, an in-depth knowledge of the relationship betweenstructures and properties of PI becomes imperative for furtherunderstanding related mechanism and improving mechanical andtribological properties. From this point of view, it is potential wayto enhance the related properties by the incorporation of somelatent cross-linkable groups, which may thermally react toprovide cross-links.

T. Huang et al. / Wear 292–293 (2012) 25–3226

Previous works on the imide oligomers have been focused onthe phenylethynl terminated oligomers [20–24]. Compared withnadic terminated oligomers, this kind of oligomers possesses goodprocessability without evolution of volatiles during the thermalcuring process [25,26]. Despite enormous reports on the perfor-mance of imide oligomers, no investigations, to date, have everbeen investigated the tribological, macro- and micro-mechanicalproperties of modified polyetherimide (PEI) prepared from phe-nylethynyl terminated etherimide (PT-EI) oligomers.

Herein, the objective of this study was aimed to investigate theeffect of phenylethynyl terminated agent on the tribological,macro- and micro-mechanical properties of PEI. 4-phenylethy-nylphthalic anhydride (4-PEPA), as the phenylethynyl terminatedagent, was added to synthesize a series of PT-EI oligomers,according to different calculated Mn. Modified PEI resins wereprepared by the thermal curing reaction of PT-EI oligomers. Forcomparison, unmodified PEI was also prepared. In order to furtherinvestigate the relationship between mechanical properties andwear resistance, mechanical properties of PEI were examined atboth macro-level and micro-level. Furthermore, appropriate cal-culated Mn of oligomers was investigated by a comparison, whichwas made in the thermal stability, mechanical and tribologicalproperties of modified PEI. In addition, worn surfaces and weardebris were characterized to understand the related tribologicalmechanisms.

Table 1M ratios, calculated Mn and intrinsic viscosity of PTAA and PAA.

Amide acid Monomers (molar ratios) Calculated Mn

(g/mol)

Za

BPADA 4,40-ODA 4-PEPA

PAA 1 1 0 n 1.21

PTAA

0.77 1 0.46 3,000 0.18

0.86 1 0.28 5,000 0.28

0.91 1 0.18 8,000 0.36

0.93 1 0.14 10,000 0.41

n The PAA solution was synthesized by the molar ratios of BPADA: 4,40-ODA as 1:1.a Measured in DMF at a concentration of 0.5 g/dL at 25 1C.

2. Experimental

2.1. Materials

2,20-bis[4-(3,4-dicarboxyphenoxy)phenyl]-propanedianhy-dride (bisphenol-A dianhydride, BPADA) was purchased fromShanghai Research Institute of Synthetic Resins and dried at120 1C under vacuum prior to use. 4,40-oxydianiline (4,40-ODA)and acetic anhydride were supplied by Sinopharm Group Chemi-cal Reagent Co., Ltd. In addition, 4,40-ODA was recrystallized fromethanol/water (1:1) prior to use. 4-phenylethynylphthalic anhy-dride (4-PEPA) was provided by Changzhou Sunlight Pharmaceu-tical Co., Ltd and purified by vacuum desiccation. N,N-dimethylformamide (DMF) and triethylamine were obtainedfrom Shanghai Qiangshun Chemical Agent Co., China. Ethanol waspurchased from Shanghai First Chemical Factory. All the reagentswere used as received.

Scheme 1. Synthesis of phenylethynyl

2.2. Synthesis of polyetheramide acid, phenylethynyl terminated

etheramide acid and etherimide oligomers

The phenylethynyl terminated etheramide acid (PTAA) oligo-mers based on BPADA, 4,40-ODA and 4-PEPA with differentcalculated Mn were synthesized, according to Scheme 1. Themolar ratios of BPADA, 4,40-ODA and 4-PEPA for the targetmolecular weight (3000, 5000, 8000 and 10000 g/mol) are shownin Table 1. The PTAA oligomers were prepared by allowing themonomers to react at room temperature (RT) in DMF for 5 h.For comparison, the polyetheramide acid (PAA) was also prepared bythe solution condensation between BPADA and 4,40-ODA with themolar ratios as 1:1, according to the previous amide acid route [12].The PTAA and PAA solutions were kept in a freezer until use.

The PT-EI oligomers were obtained via chemical synthesisroute. After the preparation of the PTAA solution, triethylamineand acetic anhydride were added as per procedure describedelsewhere [27]. The solution was allowed to stir overnight atroom temperature. The resultant mixture was then poured intohot water to yield precipitate. The solid was collected by filtra-tion, and then dried at 120 1C in vacuum for 24 h to afford anoff-white solid.

2.3. Preparation of the film specimens

The PT-EI oligomer powder was dissolved in DMF to get ahomogenous solution at a concentration of 35 wt%. The PT-EIsolution was cast onto a glass substrate, and then was heated at100, 150 and 200 1C for 1 h to get rid of volatile, followed by thethermally curing at 320 1C for 2 h. Modified PEI films wereobtained by the thermal curing reaction of the corresponding

terminated etherimide oligomers.

T. Huang et al. / Wear 292–293 (2012) 25–32 27

PT-EI solutions with different calculated Mn. For comparison,unmodified PEI (PEI-0) film was also prepared by the thermalimidization of the PAA solution. The thickness of the resultingthin films was about 0.05 mm. The details and the designations ofthe film specimens are given in Table 2.

2.4. Preparation of the friction specimens

The samples for friction and wear tests were prepared bycasting PT-EI solutions onto a stainless steel plate. Prior to coatingthe PT-EI solutions onto the steel plate, the stainless steel platewas ground with abrasive papers of various grit sizes in sequence.These specimens were thermally cured in an air convection oven,according to the given temperature procedure as described above.For comparison, unmodified PEI (PEI-0) specimen was also pre-pared. The details and designations of the friction specimens aregiven in Table 2.

2.5. Friction and wear tests

The tribological tests were conducted on a reciprocatingfriction and wear testing machine (HS-2M, Lanzhou ZhongkeKaihua Technology Development Co., Ltd). The contact schematicdiagram of the frictional couple is shown in Fig. 1. A 9Cr18stainless steel ball of diameter 6.0 mm was used as the counter-part. The roughness of the stainless ball used is 20 nm, asprovided by the supplier. The test was performed under ambientconditions (temperature: 2372 1C, relative humidity: 50710%)at a reciprocating frequency of 5 Hz and an applied load of 6 N.The length of stroke in one cycle is 8 mm. Each friction and weartest was carried out for 10 min. Before each test, the surfaces ofthe friction specimens and the counterpart ball were cleaned withacetone followed by drying.

The friction coefficient was obtained by the computer auto-matically. At the end of each test, the depth of the wear scar wasmeasured with a stylus surface profiler (Dektak 150, Veecoinstruments Inc., USA), and then the wear volume (DV, mm3) of

Table 2The details and designations of the film and friction specimens.

Designations Details

Calculated Mn (g/mol) Solutions

PEI PEI-0 n PAA

PEI-1 3,000

PT-EI

PEI-2 5,000Modified PEI

PEI-3 8,000

PEI-4 10,000

n The PAA solution was synthesized by the molar ratios of BPADA: 4,40-ODA as 1:1.

Fig. 1. Schematic diagram of the contact configuration of the reciprocating friction

and wear testing machine.

the specimen was calculated according to the following equation:

DV ¼L

2� r2arccos

r�d

r�ðr�dÞ

ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffir2�ðr�dÞ2

q� �

þ2

Z d

0p

ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffir2�ðr�dÞ2

qdð

ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffir2�ðr�dÞ2

qÞ ð1Þ

where DV is the wear volume (mm3), d the depth of the specimen(mm), L the length of stroke in one cycle (mm), and r the radius ofthe counterpart ring (mm). The wear rate (K, mm3/Nm) of thespecimen was calculated from the following equation:

K ¼DV

FtvLð2Þ

where F is the applied load (N), t the experimental duration (s),v the reciprocating frequency (Hz) and L the stroke length in onecycle (m). In this work, five replicates of friction and wear testswere carried out to minimize data scattering, and the averageresults were reported in this article.

2.6. Analysis methods

Fourier transform infrared spectroscopy (FTIR, Nexus-470,Nicolet Co., USA) was utilized to identify the ethynyl groups inthe PT-EI oligomers, and also was carried out to confirm that theethynyl groups disappear after the thermally curing procedure.

Nuclear magnetic resonance (NMR) spectra were obtained on aVarian Mercury plus 400 M instrument (Varian Co., USA) usingchloroform-d as solvents. 1H-NMR and 13C-NMR spectra wereused to confirm the success of synthesis and the existence of theethynyl groups.

Intrinsic viscosity measurement was conducted at a concen-tration of 0.5 g/dL in DMF at 25 1C, using an Ubbelohde cannon-type viscometer.

The glass transition temperature (Tg) was measured by adynamic mechanical analyzer (DMA, 242, Netzsch Co., Germany).A tension mode was employed on thin film specimens. Thetemperature scans were run from 100 to 280 1C at a heating rateof 5 1C/min. The tests were conducted at a frequency of 1 Hz,vibration amplitude being set to 120 mm, static compressivestress 4 N.

The decomposition behaviors of the PEI films were investi-gated by using thermal gravimetric analysis (TGA, Pyris 1, PerkinElmer Co., USA). The measurement was conducted in flowingnitrogen atmosphere (40 cm3/min) at a heating rate of 20 1C/min.

Tensile properties tests were carried out according to ISO 527-3:1995 standard, using a universal testing machine (CMT4104,Shenzhen Sans Testing Machine Ltd., China) at a crosshead speedof 2 mm/min. The parallel segment of the dumbbell-shapedspecimens for tensile tests was 20�4�0.05 mm. The averagevalue of five replicated measurements was adopted as the finalresult.

Nanoindentation tests were conducted with an ultra nanoin-dentation tester (CSM Instruments, Switzerlan) using a Berkovichdiamond indenter. After contacting with the surface, the indenterwas approached into the friction specimens with a constant strainrate of 0.05 s�1 until 2000 nm of depth was reached, held at themaximum load for 50 s, and then withdrawn from the surfacewith the same rate as loading. At least five indents wereperformed on each specimen and the average value was finallyadopted. The hardness was calculated by the Oliver and Pharrmethod [28,29].

In order to investigate the related tribological mechanisms, themorphologies of worn surfaces and debris were observed by usingscanning electron microscope (SEM, 5136MM, Tescan Co., Czech).All specimens were sputter coated with a thin gold layer prior toSEM observation.

Fig. 3. Typical 1H-NMR spectrum of the PT-EI oligomer in CDCl3 solvent (repre-

sented by the calculated Mn of 5000 g/mol).

T. Huang et al. / Wear 292–293 (2012) 25–3228

3. Results and discussion

3.1. Characterization of the PT-EI oligomers and PEI

The FTIR spectra of modified PEI prepared from differentcalculated Mn of oligomers and a typical PT-EI oligomer areshown in Fig. 2. In order to avoid duplication, the PT-EI oligomerwith the calculated Mn of 5000 g/mol is chosen to represent.The FTIR spectrum of unmodified PEI is also presented in Fig. 2. Itcan be found that there are several of the same characteristicabsorption bands on the spectra among the PT-EI oligomer andPEI (specimens from PEI-0 to PEI-4). The bands at about 1778 and1720 cm�1, which belong to the asymmetric and symmetricstretching vibrations of imide C¼O, are clearly observed. Thebands at about 1378 and 745 cm�1, which are assigned to thestretching and blending of imide C–N, are also detected. Inaddition, the PT-EI oligomer exhibits the characteristic bandaround 2213 cm�1, which is attributed to the stretching vibrationof ethynyl C�C. However, the characteristic band of ethynyl C�Cin the modified PEI (specimens from PEI-1 to PEI-4) disappears,which demonstrates that the PT-EI oligomers have been fullythermally cured after the multistage heating schedule. Thus, fromcharacteristic peaks discussed above, it was roughly proved thatthe expected chemical structures have been obtained according toScheme 1.

Figs. 3 and 4 show 1H-NMR and 13C-NMR spectra of the PT-EIoligomer (represented by the calculated Mn of 5000 g/mol),respectively. It can be seen that the PT-EI oligomer has beenprepared successfully, according to the 1H-NMR spectrum and13C-NMR spectrum. Moreover, the 13C-NMR spectrum shows theintegrals of the carbons in the backbone (–CH3, d¼31 ppm) andthe end-capping group (–C�C, d¼88 and 94 ppm). Thus it isfurther demonstrated that the PT-EI oligomers have been synthe-tized successfully according to Scheme 1.

Molecular weight control was achieved by utilizing 4-PEPA asan end-capping agent. As shown in Table 1, the intrinsic viscosityof the PAA and PTAA solutions was measured by Ubbelohde

Fig. 2. FTIR spectra of the PEI and PT-EI oligomer (represented by the calculated

Mn of 5000 g/mol).

Fig. 4. Typical 13C-NMR spectrum of the PT-EI oligomer in CDCl3 solvent

(represented by the calculated Mn of 5000 g/mol).

viscometer at a concentration of 0.5 g/dL in DMF at 25 1C. It canbe found that different calculated Mn is corresponding to differentintrinsic viscosity, which is consistent with previous reports ofother researchers [23,30]. Therefore, it can be proved indirectlythat the PT-EI oligomers with different calculated Mn have beensynthesized by the phenylethynyl terminated agent.

3.2. Macro-mechanical properties of PEI

Fig. 5 shows the effect of phenylethynyl terminated agent onthe tensile strength and elongation at break of PEI. As shown inTable 1, the PT-EI oligomers with different calculated Mn wereprepared according to different addition of phenylethynyl termi-nated agent. The PEI-0 and modified PEI (specimens from PEI-1 toPEI-4) have been prepared, respectively, according to Table 2.It can be clearly seen that the incorporation of phenylethynylterminated agent significantly enhances the tensile strength ofPEI and causes a reduction in elongation at break. It is worthnoting that the highest tensile strength is obtained at thecalculated Mn of 5000 g/mol and this is ca. 20% higher than thatof the PEI-0 (87.14 MPa).

Fig. 5. Effect of phenylethynyl terminated agent on the tensile strength and

elongation at break of PEI.

Fig. 6. Effect of phenylethynyl terminated agent on the loading-hold-unloading

curves and microhardness of PEI.

Table 3Data based on TGA and DMA of PEI.

Samples T5a (1C) T10

b (1C) Rwc (%) Tg (1C)

PEI-0 546.9 558.1 51.3 216.7

PEI-1 537.0 549.9 54.1 224.2

PEI-2 543.3 555.7 57.6 238.2

PEI-3 541.2 552.4 54.6 231.7

PEI-4 547.5 559.6 58.2 228.3

a T5: the temperature at 5 wt% of weight loss was recorded by TGA with a

heating rate of 20 1C/min under N2 atmosphere.b T10: the temperature at 10 wt% of weight loss was recorded by TGA with a

heating rate of 20 1C/min under N2 atmosphere.c Rw: residual weight retention was recorded by TGA at 800 1C with a heating

rate of 20 1C/min under nitrogen atmosphere.

T. Huang et al. / Wear 292–293 (2012) 25–32 29

When the calculated Mn is above 5000 g/mol, however, thetensile strength of modified PEI decreases slightly and theelongation at break improves. This implies that the calculatedMn of oligomers plays an important role in the macro-mechanicalproperties of PEI. The difference in the changes in the tensilestrength as a function of calculated Mn of oligomers can bepossibly explained by the difference in the backbone compositionand crosslink density what we will discuss below. Apparently, it isfound that appropriate calculated Mn of oligomers can markedlyimprove the tensile strength of PEI.

3.3. Micro-mechanical properties of PEI

Fig. 6 shows the effect of phenylethynyl terminated agent onthe typical loading-hold-unloading curves and microhardness ofPEI. On loading, the force increases with increasing depth. It canbe seen that the loading-hold-unloading curve of PEI obviouslyshifts upwards because of the incorporation of phenylethynylterminated agent, indicating that modified PEI (specimens fromPEI-1 to PEI-4) specimens’ resistance to indentation significantlyincreases comparing with PEI-0. It is well noting that thedeformations of modified PEI specimens are similar after unload-ing and are less than that of PEI-0. While the curves of modifiedPEI initially shift upwards with increasing calculated Mn, and then

go down clearly when the calculated Mn is above 5000 g/mol. Thevalues of microhardness of PEI show a signficant increase from397.7 to 548.2 MPa because of the incorporation of phenylethynylterminated agent. Obviously, the above results support the ideathat the appropriate calculated Mn of oligomers by the addition of4-PEPA causes the improvement in micro-mechanical propertiesof PEI.

Surprisingly, the change trend of microhardness among the PEIis completely the same as that of tensile strength. It indicates thatthe influence of the incorporation of phenylethynyl terminatedagent on the macro- and micro-mechanical properties displaysthe same effect, which is due to an essential factor that thechanges in the macromolecular structures of PEI.

3.4. Thermal properties of PEI

The thermal properties of PEI, which were thermally cured,were evaluated by DMA and TGA. The thermal data of PEI basedon TGA and DMA are summarized in Table 3. It is obviously foundthat the glass transition temperature (Tg) of modified PEI (speci-mens from PEI-1 to PEI-4) is enhanced significantly in differentlevels, compared with that of the PEI-0. This phenomenon may bewell explained that the incorporation of the phenylethynylterminated agent offers some latent cross-linkable groups, whichmakes the polymer backbone more rigid after the thermallycuring process.21 Moreover, the PEI-2 shows the highest Tg of238.2 1C, which is ca. 20 1C higher than that of the PEI-0(216.7 1C). Combining with macro- and micro-mechanical char-acterization above, it can be inferred that the optimal calculatedMn of the PT-EI oligomers can be 5000 g/mol.

Comparing the weight retentions between the PEI-0 andmodified PEI, it is found that the incorporation of phenylethynylterminated agent improves the weight retentions of modified PEI,and these values are higher than 54.1%. In addition, the modifiedPEI specimens exhibit excellent thermal stability with 5 and10 wt% of weight loss in the range of 537–547.5 and of 549.9–559.6 1C in nitrogen atmosphere, respectively. This indicates thatmodified PEI specimens still keep excellent thermal stability afterthe incorporation of phenylethynyl terminated agent.

3.5. Friction and wear properties of PEI

The friction coefficient and wear rate of PEI specimens wereconducted under a load of 6 N, a reciprocating frequency of 5 Hzand dry condition. Fig. 7 presents the effect of phenylethynylterminated agent on the friction coefficient and wear rate of PEIspecimens.

Compared with the PEI-0, the wear rate of modified PEI(specimens from PEI-1 to PEI-4) shows lower values clearly. Whatis more, the wear rate of the PEI-2 prepared from the oligomer

T. Huang et al. / Wear 292–293 (2012) 25–3230

with the calculated Mn of 5000 g/mol reaches the minimum valueand the lowest wear rate is 3.8�10�5 mm3/Nm, which meansthat the wear resistance of the PEI-2 is ca. five times more thanthat of the PEI-0. The degree of substantial improvement in the

Fig. 7. Effect of phenylethynyl terminated agent on the friction coefficient and

wear rate of PEI.

Fig. 8. SEM micrographs of worn surfaces and wear debris of PEI specimens (magn

(a, b) PEI-0, (c, d) PEI-1, (e, f) PEI-2, (g, h) PEI-3 and (j, k) PEI-4

wear resistance of the PEI is similar to that obtained in someprevious studies [11–15]. In addition, it can be seen that the wearrate of modified PEI initially decreases sharply and then increasesslightly with increasing calculated Mn. Therefore, the optimumcalculated Mn of oligomers is 5000 g/mol considering the wearresistance of PEI.

As shown in Fig. 7, the incorporation of phenylethynyl termi-nated agent obviously causes to increase the friction coefficient ofPEI. It is also found that the friction coefficient is 0.51 at themaximum value when the calculated Mn reaches 5000 g/mol.

3.6. Worn surfaces and wear debris observations

To make clear the tribological mechanisms of PEI, wornsurfaces and wear debris of the PEI-0 and modified PEI wereinvestigated with SEM.

The SEM images of the worn surfaces of the PEI-0 and modifiedPEI under experimental conditions are shown in Fig. 8a, c, e, g andj. It can be seen that the wear width of modified PEI (specimensfrom PEI-1 to PEI-4) is much smaller compared with that of thePEI-0 (0.66 mm). This demonstrates that the incorporation ofphenylethynyl terminated agent can obviously enhances the wearresistance of PEI. Furthermore, it is also found that the wear width

ification: 200; load: 6 N; reciprocating frequency: 5 Hz; test duration: 10 min).

T. Huang et al. / Wear 292–293 (2012) 25–32 31

of modified PEI first reduces and then increases with increasingcalculated Mn. The wear width is at its smallest value when thecalculated Mn is 5000 g/mol. Apparently, The change trend ofwear rate as discussed above is well consistent with the change inthe wear width.

In addition, it can be seen in Fig. 8 that the worn surface of thePEI-0 is relatively smoother than those of modified PEI (speci-mens from PEI-1 to PEI-4) and shows signs of light adhesion andabrasive wear. However, the worn surfaces of modified PEI arerough, displaying deep scuffed and furrowed marks. Moreover, itis obviously seen that fatigue deformations on the worn surfacesof modified PEI are severer to some extent compared with that ofthe PEI-0, indicating that the type of wear changed from abrasivewear into fatigue wear. Surprisingly, the change trend of surfaceroughness of worn surfaces (Fig. 8a, c, e, g and j) is roughlyconsistent with the change trend of friction coefficient of thecorresponding PEI. This might be inferred that the surface rough-ness of worn surface is one of the most important factors thatgovern the friction coefficient of materials.

Debris is a product of friction experiments. Therefore, itsanalysis can be helpful to understand the related friction andwear mechanism. The SEM images of wear debris of PEI speci-mens are given in Fig. 8b, d, f, h and k. Both wear debris of the PEI-0 and modified PEI almost consists of flakes. Nevertheless, debrisof modified PEI (specimens from PEI-1 to PEI-4) is considerablysmaller than that of the PEI-0. Furthermore, it is also found thatthe size of debris first decreased with increasing calculated Mn,however, the debris size of modified PEI increases when thecalculated Mn exceeds 5000 g/mol. it is worth noting that thewear debris of the PEI-2 turns to be small sized platelets orparticles. The size of the debris agrees well with the wearresistance of the corresponding PEI specimens, considering therelationship between the debris size and wear rate.

3.7. Discussion

It is worth pointing out that mechanical and tribologicalproperties of a material is dependent on a number of factors,which includes macromolecule structures, composition forblends, processing, surface tension, interfacial adhesion, etc[31–41]. In the case of modified PEI in this study, macromoleculestructures were modified by the incorporation of phenylethynylterminated agent. The PT-EI oligomers are rich in ethynyl groups(cross-linkable groups) (Fig. 2). Upon the thermal curing reaction,the ethynyl groups have undergone a complex reaction involvingchain extension, branching and crosslinking without the evolu-tion of volatile by-products, which makes the polymer backbonemore rigid [21–23]. Although the definite structure of modifiedPEI (specimens from PEI-1 to PEI-4) is not well known, modifiedPEI specimens exhibit an excellent combination of properties thatinclude higher Tg, tensile strength and microhardness, comparedwith PEI-0. Moreover, there definitely exists a close relationshipbetween mechanical properties and wear resistance of PEI. Inother words, the improvement in macro- and micro-mechanicalproperties of PEI contributes to obtain better wear resistance.

In addition, SEM observations of wear debris (Fig. 8) indicatethat modified PEI with higher tensile strength and microhardnessshows relatively small flakes or particles. According to Ratner-Lancaster [42,43], it could be explained that the material may bemore difficult to be peeled off because of more excellent mechan-ical properties, and hence that results in a lower wear.

By making further comparison among the properties of mod-ified PEI (specimens from PEI-1 to PEI-4), it is apparently foundthat the best performances, including Tg, mechanical and tribologicalproperties, are obtained when the calculated Mn is 5000 g/mol.In other words, the calculated Mn which is greater or less than

5000 g/mol goes against the improvement of the performance ofPEI. It may be explained as follows: First, the PT-EI oligomerswould be rich in ethynyl groups when the calculated Mn isrelatively small. The backbone composition of the PEI mightmainly consist of carbon chain (alkene C¼C) after the thermalcuring. The carbon backbone is generally much weaker than thearomatic backbone. Second, there would be poor in cross-linkablegroups when the calculated Mn is relatively large. That wouldresult in low crosslink density after the thermal curing, which isalso demonstrated by Hergenrother et al. [44]. Therefore thebackbone composition and crosslink density are important fac-tors, which influence the performances of PEI.

4. Conclusion

Modified PEI resins were prepared by the thermal curingreaction of PT-EI oligomers. The incorporation of 4-PEPA wasutilized to synthesize a series of PT-EI oligomers, according todifferent calculated Mn. The tribological properties of PEI speci-mens were investigated by a reciprocating friction and weartesting machine. The macro- and micro-mechanical propertieswere evaluated to further understand the relationship betweenmechanical and tribological properties. The friction and wearmechanism was discussed by analyzing the worn surfaces andwear debris. The major conclusions can be summarized asfollows:

(1)

The incorporation of phenylethynyl terminated agentincreases Tg, tensile strength and microhardness of PEI.Modified PEI prepared from the oligomer with the calculatedMn of 5000 g/mol shows the highest tensile strength, which isca. 20% higher than that of the PEI-0. Moreover, the weightretentions of modified PEI are improved gently comparedwith that of the PEI-0.

(2)

The incorporation of phenylethynyl terminated agent contri-butes to a significant improvement on the tribological proper-ties. The wear rate of modified PEI specimens initiallydecreases sharply and then increases slightly with increasingcalculated Mn. it is worth noting that modified PEI preparedfrom the oligomer with the calculated of 5000 g/mol showsthe best wear resistance, which is ca. five times more thanthat of unmodified PEI. SEM observations of worn surfacesshow that type of wear changed from abrasive wear intofatigue wear.

(3)

The improvement in macro- and micro-mechanical propertiesof modified PEI contributes to obtain better wear resistance.

(4)

The calculated Mn of oligomers is one of significant factorswhich affect Tg, mechanical and tribological properties.Apparently, modified PEI shows the best performance whenthe calculated Mn is 5000 g/mol.

References

[1] P. Samyn, G. Schoukens, F. Verpoort, J. Van Craenenbroeck, P. De Baets,Friction and wear mechanisms of sintered and thermoplastic polyimidesunder adhesive sliding, Macromolecular Materials and Engineering 292(2007) 523–556.

[2] T. Huang, R.G. Lu, C. Su, H.N. Wang, Z. Guo, P. Liu, Z.Y. Huang, H.M. Chen,T.S. Li, Chemically modified graphene/polyimide composite films based onutilization of covalent bonding and oriented distribution, ACS AppliedMaterials & Interfaces 4 (2012) 2699–2708.

[3] S.K. Sinha, B.J. Briscoe, Polymer tribology, Landon, Imperial College Press,2009.

[4] P. Samyn, G. Schoukens, P. De Baets, Micro- to nano-scale surface morphologyand friction response of tribological polyimide surfaces, Applied SurfaceScience 256 (2010) 3394–3408.

T. Huang et al. / Wear 292–293 (2012) 25–3232

[5] D.H. Buckley, Friction and Wear Characteristics of Polyimide and FilledPolyimide Compositions in Vacuum, NASA TN D-3261, Washington D.C, 1966.

[6] P.H. Cong, T.S. Li, X.J. Liu, Effect of temperature on friction and wear ofpolyimide, Tribology 17 (1997) 220.

[7] X.R. Zhang, X.Q. Pei, Q.H. Wang, Friction and wear properties of polyimidematrix composites reinforced with short basalt fibers, Journal of AppliedPolymer Science 111 (2009) 2980–2985.

[8] P. Samyn, G. Schoukens, Tribological properties of PTFE-filled thermoplasticpolyimide at high load, velocity, and temperature, Polymer Composites 30(2009) 1631–1646.

[9] P. Samyn, G. Schoukens, P. De Baets, Evaluation of morphology and depositson worn polyimide/graphite composite surfaces by contact-mode AFM, Wear270 (2010) 57–72.

[10] X.R. Zhang, X.Q. Pei, Q.H. Wang, Friction and wear studies of polyimidecomposites filled with short carbon fibers and graphite and micro SiO2,Materials & Design 30 (2009) 4414–4420.

[11] Y.J. Shi, L.W. Mu, X. Feng, X.H. Lu, The tribological behavior of nanometer andmicrometer TiO2 particle-filled polytetrafluoroethylene/polyimide, Materials& Design 32 (2011) 964–970.

[12] S.Q. Lai, T.S. Li, F.D. Wang, The effect of silica size on the friction and wearbehaviors of polyimide/silica hybrids by sol–gel processing, Wear 262 (2007)1048–1055.

[13] J. Kim, H. Im, M.H. Cho, Tribological performance of fluorinated polyimide-based nanocomposite coatings reinforced with PMMA-grafted-MWCNT,Wear 271 (2011) 1029–1038.

[14] Q.H. Wang, X.R. Zhang, X.Q. Pei, A synergistic effect of graphite and nano-CuOon the tribological behavior of polyimide composites, Journal of Macromo-lecular Science, Part B 50 (2011) 213–224.

[15] S.Q. Lai, Y Li, T.S. Li, X.J. Liu, R.G. Lv, An investigation of friction and wearbehaviors of polyimide/attapulgite hybrid materials, Macromolecular Mate-rials and Engineering 290 (2005) 195–201.

[16] N. Satyanarayana, K.S. Skandesh Rajan, S.K. Sinha, L. Shen, Carbon nanotubereinforced polyimide thin-fire for high wear durability, Tribology Letters 27(2007) 181–188.

[17] P.S.G. Krishnan, A.E. Wisanto, S. Osiyemi, C. Ling, Synthesis and properties ofBCDA-based polyimide-clay nanocomposites, Polymer International 56(2007) 787–795.

[18] M.R. Chitsaz-Zadeh, N.S. Eiss, Friction and wear of polyimide thin films, Wear110 (1986) 359–368.

[19] J.S. Tian, H.Y. Wang, H.Z. Huang, R.G. Lu, P.H. Cong, X.J. Liu, T.S. Li, Investiga-tion on tribological properties of fluorinated polyimide, Journal of Macro-molecular Science, Part B 49 (2010) 791–801.

[20] Y. Yang, L. Fan, M. Ji, S.Y. Yang, Phenylethynylnaphthalic endcapped imideoligomers with reduced cure temperatures, European Polymer Journal 46(2010) 2145–2155.

[21] Y. Yang, L. Fan, X.M. Qu, M. Ji, S.Y. Yang, Fluorinated phenylethynyl-terminated imide oligomers with reduced melt viscosity and enhanced meltstability, Polymer 52 (2011) 138–148.

[22] H.J. Sun, H.T. Huo, H. Nie, S.Y. Yang, L. Fan, Phenylethynyl terminatedoligoimides derived from 3,30 ,4,40-diphenylsulfonetetracarboxylic dianhy-dride and their adhesive properties, European Polymer Journal 45 (2009)1169–1178.

[23] B Tan, V. Vasudevan, Y.J. Lee, S. Gardner, R.M. Davis, T. Bullions, A.C. Loos,H. Parvatareddy, D.A. Dillard, J.E. Mcgrath, J. Cella, Design and characteriza-tion of thermosetting polyimide structural adhesive and composite matrixsystems, Journal of Polymer Science Part A 35 (1997) 2943–2954.

[24] C.D. Simone, D.A. Scola, Modification of PETI-5K imide oligomers: effect onviscosity, High Performance Polymers 15 (2003) 473–501.

[25] X.M. Fang, D.F. Rogers, D.A. Scola, M.P. Stevens, A study of the thermal cure ofa phenylethynyl-terminated imide model compound and a phenylethynyl-terminated imide oligomer (PETI-5), Journal of Polymer Science Part A 36(1998) 461–470.

[26] P.M. Hergenrother, J.G. Smith Jr, Chemistry and properties of imide oligomersend-capped with phenylethynylphthalic anhydrides, Polymer 35 (1994)4857–4864.

[27] S. Tamai, T. Kuroki, A. Shibuya, A. Yamaguchi, Synthesis and characterization ofthermally stable semicrystalline polyimide based on 3,40-oxydianiline and3,30 ,4,40-biphenyltetracarboxylic dianhydride, Polymer 42 (2001) 2373–2378.

[28] W.C. Oliver, G.M. Pharr, An improved technique for determining hardnessand elastic modulus using load and displacement sensing indentationexperiments, Journal of Materials Research 7 (1992) 1564–1583.

[29] W.C. Oliver, G.M. Pharr, Measurement of hardness and elastic modulus byinstrumented indentation: advances in understanding and refinements tomethodology, Journal of Materials Research 19 (2004) 3–19.

[30] G.X. Su, J.Y. Wang, G.H. Li, X.G. Jian, Synthesis and properties of solublecopolyimides containing phthalazinone moieties for flexible copper-cladlaminates, High Performance Polymers 22 (2010) 930–944.

[31] L.N. Liu, A.J. Gu, Z.P. Fang, L.F. Tong, Z.B. Xu, The effects of the variations ofcarbon nanotubes on the micro-tribological behavior of carbon nanotubes/bismaleimide nanocomposite, Composites Part A 38 (2007) 1957–1964.

[32] P. Guo, H. Song, X. Chen, Interfacial properties and microstructure of multi-walled carbon nanotubes/epoxy composites, Materials Science and Engineer-ing A 517 (2009) 17–23.

[33] P. Liu, R.G. Lu, T. Huang, P.H. Cong, S.S. Jiang, T.S. Li, Tensile and tribologicalproperties of polytetrafluroethylene homocomposites, Wear 289 (2012)65–72.

[34] P. Liu, T. Huang, R.G. Lu, T.S. Li, Tribological properties of modified carbonfabric/polytetrafluoroethylene composites, Wear 289 (2012) 17–25.

[35] W. Brostow, J.L. Deborde, M. Jaklewicz, P. Olszynski, Tribology with emphasison polymers: friction, scratch resistance and wear, Journal of MaterialsEducation 25 (2003) 119–132.

[36] T. Huang, R.G. Lu, H.Y. Wang, Y.N. Ma, J.S. Tian, T.S. Li, Investigation on thetribological properties of POM modified by nano-PTFE, Journal of Macro-molecular Science, Part B 50 (2011) 1235–1248.

[37] W. Brostow, P.E. Cassidy, J. Macossay, D. Pietkiewicz, S. Venumbaka, Slidingwear, viscoelasticity, and brittleness of polymers, Polymer International 52(2003) 1498–1505.

[38] T. Huang, R.G. Lu, Y.N. Ma, P. Liu, T.S. Li, Study on the friction and sliding wearbehavior of hybrid polytetrafluoroethylene/Kevlar fabric composites filledwith polyphenylene sulfide, Journal of Macromolecular Science, Part B 51(2012) 109–124.

[39] W. Brostow, H.E. Hagg Lobland, M. Narkis, Sliding wear, viscoelasticity, andbrittleness of polymers, Journal of Materials Research 21 (2006) 2422–2428.

[40] T. Huang, R.G. Lu, Y.N. Ma, P. Liu, H.M. Chen, Z.Y. Huang, T.S. Li, Surfacemodification by air-plasma treatment and its effect on the tribologicalbehavior of hybrid fabric-polyphenylene sulfide composites, Journal ofMacromolecular Science, Part B 51 (2012) 1011–1026.

[41] O. Joachim, S. Katja, Bulk nanostructured materials, in: M.J. Zehetbauer,Y.T. Zhu (Eds.), 3D Images of Materials Structures, Wiley-VCH Verlag GmbH &Co. KGaA, Weinheim, 2009.

[42] G.W. Stachowiak, A.W. Batchelor, Engineering Tribology, Elsevier Butter-worth-Heinemann, Amsterdam, 2005.

[43] I.M. Hutchings, Tribology: Friction and Wear of Engineering Materials, Else-vier Limited, London, 1992.

[44] P.M. Hergenrother, J.W. Connell, J.G. Smith Jr, Phenylethynyl containingimide oligomers, Polymer 41 (2000) 5073–5081.