polymers Article
Mechanical and Unlubricated Sliding Wear Properties of Nitrile Rubber Reinforced with Micro Glass Flake Yanbao Guo 1,2, *, Hai Tan 1 , Zhiqiang Cao 1,3 , Deguo Wang 1,2, * and Siwei Zhang 1 1 2 3
*
College of Mechanical and Transportation Engineering, China University of Petroleum, Beijing 102249, China;
[email protected] (H.T.);
[email protected] (Z.C.);
[email protected] (S.Z.) Beijing Key Laboratory of Process Fluid Filtration and Separation, Beijing 102249, China China Petroleum Pipeline Engineering Co., Ltd., Langfang 065000, China Correspondence:
[email protected] (Y.G.);
[email protected] (D.W.); Tel.: +86-10-89733727 (Y.G.)
Received: 16 May 2018; Accepted: 22 June 2018; Published: 26 June 2018
Abstract: In this study, the filled nitrile rubber (NBR) was prepared with micro glass flake (GF). The tribological behaviors of filled NBR were tested by a ball-on-disk tribometer. Material properties such as glass transition temperature (Tg ), fracture energy, tensile strength and dispersity of GF filler were also investigated. The results showed that the coefficient of friction (COF) of NBR reduced and the wear-resistant enhanced with the GF filler. Compared to unfilled NBR, the COF of filled NBR suffered a maximal drop percent (about 21.1%) at a rotation speed of 100 rpm and normal load of 1.5 N. Mechanical and wear behaviors were dependent on the interfacial performance of filler in the rubber matrix. Filler with smaller size was more conducive to enhance the interfacial strength of the polymer matrix. That can increase the interfacial strength of filler and benefits to improve the anti-wear behavior of rubber. Keywords: micro glass flake; nitrile rubber; wear resistance; friction and wear
1. Introduction As a preferred material for many applications, nitrile rubber (NBR) is widely used in various industries including sealing and anti-wear components because of its good properties such as physical, chemical and thermostability [1–3]. Accordingly, a great deal of scientific researches have been focused on the properties of NBR and its applications. However, according to the failure statistics, the wear of rubber is one of the foremost contributors to the failure of polymer parts [4]. It was pointed out that wear of rubber was the main cause of these failures, which account for 60~80% [5]. Furthermore, the wear of rubber is still an inevitable problem for the seal components and moving components under rough contact and harsh terms, especially in a moving system. This could lead to an enormous economic loss [6]. For example, in screw pump systems, the wear in elastomer stator is a time-consuming and costly problem [7]. A lot of researches have been carried out to investigate the wear behaviors of rubber, for instance, Cao et al. [8] prepared NBR with different cross-linking densities and studied the cross-linking heterogeneity effects on the wear behaviors of NBR. It was found that the cross-linked heterogeneity shows a critical factor in wear properties of unfilled NBR. Furthermore, the operating conditions and its structure factors also influence the wear behaviors of polymer. Chen et al. [9] studied the friction and wear performances of polymers sliding against GCr15 and 316 steels under sea water lubricant condition. From the test results, it was found that the tribological properties of a polymer with the lubrication of sea water were not only influenced by the properties of the polymer itself, but also the corrosion and lubrication of the aqueous medium. Shen et al. investigated the tribological behaviors of NBR in two-body abrasion. The results indicated that the size of abrasive showed a significant effect on the tribological and mechanical properties of Polymers 2018, 10, 705; doi:10.3390/polym10070705
www.mdpi.com/journal/polymers
Polymers 2018, 10, 705
2 of 17
the NBR [10]. Guo and co-authors [11] prepared bimodal NBR and evaluated the friction and wear performances of rubber. The test results showed that the pattern of the rubber network could be transferred from the elastomeric to the thermosetting resin. In addition, to improve the wear performance of rubber and meet the demands of rubber application, more and more polymer composites with enhanced behaviors are studied including many types of fillers-reinforced polymers [12]. It is beneficial to satisfy the urgent demands of environmental protection, energy saving, cost reducing, and emission reduction [13]. Therefore, the research of polymer composite materials with filler has been one of the most popular research fields [14]. Gatos and Karger-Kocsis [15] studied the influences of the aspect ratio of layered filler on the oxygen permeation and mechanical behaviors of hydrogenated nitrile rubber (HNBR) filled with organophilic layered silicate. Expanded graphite (EG)- and modified expanded graphite (MEG)-filled composite polymers were reported to be fabricated based upon natural rubber with or without carbon black [16]. EG/MEG added natural rubber compounds, whether in the presence of carbon black showed comprehensive enhancement, such as in the thermal properties, dynamic mechanical, cure characteristics, and so on. Furthermore, fibrillar structure fillers such as fibrillar silicate were also used to load into the polymer in order to enhance its mechanical, tribological and thermal performances [17,18]. The investigation indicated that the dispersion of the fibrillar silicate filler and interfacial strength between the rubber and fibrillar silicate were able to be improved by modifying the fibrillar silicate surface, thus the polymer’s mechanical performances were improved [19]. In the previous investigations, the tribological properties of NBR which contains hollow glass beads (HGB) showed that the coefficient of friction (COF) of NBR was reduced and reinforced wear-resistant was obtained by adding the HGB filler into the rubber matrix. Furthermore, the contact interface between the steel ball and NBR could also be improved by the chips of HGB, and the extension of curl-shaped worn of rubber surface was impeded because of the chips on the frictional interface [20]. According to the previous researches, glass flake (GF) is known as a high aspect-ratio reinforcing additive which appears in many commercial applications, such as anti-corrosion [21,22], fire protection [23], and denture materials [24]. The micro flake is a modified ‘C’ glass composition which is used as filler to improve the mechanical performances of polymer composites. For instance, the glass flake-reinforced polypropylene (PP) was prepared for analyzing the impact resistance. Results indicated that the dynamic fracture toughness of the glass flake-filled PP composite increased obviously, and it was intensely rate-dependent with the adding of glass flakes [25]. Glass flakes can substantially improve the denture base acrylic material polymethyl methacrylate (PMMA) in fracture toughness for a 69% increase [24]. Klaas et al. [26] investigated the tribological behaviors of glass-filled polytetrafluoroethylene (PTFE), and the results showed that PTFE composites filled with different glass fillers behaved in different abilities for forming transferred film on the counterface, and then affected their sliding wear. Therefore, it has attracted increasing attention in many industrial polymer fields to use glass flake as a reinforcing agent. In this study, the NBR-vulcanized specimens were prepared with different sizes micro glass flakes in accordance with a certain formula proportion. GF were blended into nitrile rubber step by step through a two-roll mill and the sulfuration process. In order to characterize the NBR vulcanizate, universal tests were carried out including dynamic mechanics analysis (DMA) and differential scanning calorimetry (DSC). The tribological properties of GF-reinforced NBR have been examined using a ball-on-disk frictional instrument. Furthermore, the micro wear tracks were observed by scanning electron microscope (SEM) after the friction test. 2. Experiment Section 2.1. Materials and Preparation of NBR Specimens Nitrile rubber (N215SL, contain with 48% of acrylonitrile, JSR Corporation, Tokyo, Japan) was used as basis material. Sulfur (99.5% purity, Langfang, China) was purchased from Kezhan Chemical
Polymers 2018, 10, x FOR PEER REVIEW
3 of 17
2. Experiment Section Polymers 2018, 10,and 705 Preparation 2.1. Materials
of NBR Specimens
3 of 17
Nitrile rubber (N215SL, contain with 48% of acrylonitrile, JSR Corporation, Tokyo, Japan) was used as basis material. (99.5%agent. purity,The Langfang, China)flakes was purchased fromwere Kezhan Chemical Company and used asSulfur the curing micro glass (GF, ‘C’ glass) supplied by Company and used Pty as the curing agent. The Australia). micro glassThe flakes (GF, ‘C’ glass) were by Glassflake Australia Ltd. (Perth, Western chemical constituents ofsupplied micro glass Glassflake Australia Pty 1.Ltd. (Perth, Western Thelengths chemical of filler microinglass flake are given in Table Three styles of GF Australia). with different areconstituents employed as this flake are given in Table 1. Three styles GF with different lengths are employed as fillerin inthese this study. study. The micro morphology of theof three styles GF with different sizes are shown SEM The micro morphology of the three styles GF styles with different sizes in 35 these images images (Figure 1). The average lengths of three GF are 425 um,are 300shown um and um,SEM respectively. (Figure 1). The average lengths of three styles GF are 425 um, 300 um and 35 um, respectively. The nominal thickness of the glass flake is around 5 ± 2 µm. Thereafter, the filled NBR with threeThe GF nominal thickness of the GF-M glass flake is around 5 ± 2 μm. Thereafter, the and filled35NBR three GF are are represented as GF-B, and GF-S for lengths of 425 µm, 300 µm µm, with respectively. represented as GF-B, GF-M and GF-S for lengths of 425 μm, 300 μm and 35 μm, respectively.
SiO2
Table 1. Chemical constituents of micro glass flake (‘C’ glass). Table 1. Chemical constituents of micro glass flake (‘C’ glass). K2 O B2 O3 ZnO Na2 O MgO CaO Al2 O3
SiO2 K2O 64–70% 0–3% 64–70% 0–3%
B2O3 3–8% 3–8%
ZnO 0–5% 0–5%
Na2O 11–18% 11–18%
(a)
(b)
Figure 1. Cont.
MgO 1–4% 1–4%
CaO 3–7% 3–7%
Al2O3 0–5% 0–5%
TiO2
TiO2 0–3%
0–3%
Polymers 2018, 10, 705
4 of 17
Polymers 2018, 10, x FOR PEER REVIEW
4 of 17
(c)
Figure 1. SEM images of three type micro glass flakes: (a) GF-B, (b) GF-M and (c) GF-S. Figure 1. SEM images of three type micro glass flakes: (a) GF-B, (b) GF-M and (c) GF-S.
Moreover, N-Cyclohexy-2-benzothiazole sulfonamide (CBS) was used as the vulcanization Moreover, N-Cyclohexy-2-benzothiazole sulfonamide (CBS) was used asingredients the vulcanization accelerator as well as tetramethyl thiuram disulfide (TMTD). Additionally, other such as accelerator as well stearic as tetramethyl thiuram disulfide (TMTD).grade Additionally, other ingredients such as zinc oxide (ZnO), acid, and sulfur were commercial chemicals. These chemicals were zinc (ZnO), stearic acid, and sulfur were commercial grade These chemicals were usedoxide as supplied without further processing. Table 2 shows thechemicals. specific formula proportion of used as supplied without further processing. Table 2 shows the specific formula proportion of samples. samples. Vulcanizates of NBR were fully mixed by a two-roll mill. Afterwards, a hydraulic press was Vulcanizates of NBR the wereNBR fullysamples mixed by a two-roll mill. Afterwards, hydraulic press was utilized to prepare under a pressure of 20 MPaaand temperature of utilized 155 °C. ◦ C. Additionally, to prepare the NBR samples under a pressure of 20 MPa and temperature of 155 Additionally, a moving die rheometer (GoTech-M2000A, Taiwan, China) was used to identify the aoptimal movingcure die time. rheometer (GoTech-M2000A, Taiwan, China) was used to identify the optimal cure time. Table 2. Formula of Table 2. Formula of nitrile nitrile rubber rubber (NBR) (NBR) vulcanizates vulcanizates reinforced reinforced with with micro micro glass glass flake flake (GF). (GF).
Ingredient Stearic CBS Ingredient NBRNBR Stearic AcidAcidZnOZnO CBS PHRPHR 100 100
1
1
5
5
1.5 1.5
TMTD TMTD
Sulfur Sulfur
0.2 0.2
1.7 1.7
GF GF 1515
2.2. Experimental Instruments and Measurements 2.2. Experimental Instruments and Measurements 2.2.1. Mechanical Tests 2.2.1. Mechanical Tests The universal universal test test machine machine (WDTII-20, (WDTII-20, Guangzhou, Guangzhou, China) was used used for for evaluating evaluating uniaxial uniaxial The China) was tensile properties of rubber specimens. In this study, according to the standard of ISO 37: 2011, the tensile properties of rubber specimens. In this study, according to the standard of ISO 37: 2011, crosshead velocity was 500 mm/min. Before the test, the two clamp holders of instrument were set the crosshead velocity was 500 mm/min. Before the test, the two clamp holders of instrument were set with an an initial initial space space of of 36 36 mm. mm. Then, with Then, the the average average value value can can be be obtained obtained from from results results of of five five different different samples. Furthermore, the fracture energy (G c) of specimen was measured with the trouser tear test samples. Furthermore, the fracture energy (Gc ) of specimen was measured with the trouser tear test mode which whichwas wasalso also performed by universal the universal test machine. The crosshead velocity was 5 mode performed by the test machine. The crosshead velocity was 5 mm/min. mm/min. According to Equation (1), the fracture energy, G c, of NBR can be calculated according to According to Equation (1), the fracture energy, Gc , of NBR can be calculated according to following following[27]. equation [27]. equation 2F
G = 2F G cc = t t
(1) (1) where, Gc is the fracture energy of rubber specimens. F is the tear force. t is represented as the thickness of the test where, Gc piece. is the fracture energy of rubber specimens. F is the tear force. t is represented as the thickness of the test piece. 2.2.2. Differential Scanning Calorimetry (DSC)
2.2.2.DSC Differential Scanning Calorimetry (DSC) instrument (NETZSCH, 204F1, Bavaria, Germany) was used to evaluate the DSC ◦ C, and the rise rate was characterization. The test temperature wasBavaria, ranged from −70 ◦ Cwas to 20 DSC instrument (NETZSCH, 204F1, Germany) used to evaluate the DSC characterization. The test temperature was ranged from −70 °C to 20 °C, and the rise rate was 10
Polymers 2018, 10, 705
5 of 17
Polymers 2018, 10, x FOR PEER REVIEW
5 of 17
10 ◦ C/min. Furthermore, the glass transition temperature (Tg ) of specimens can be determined from °C/min. Furthermore, the glass transition temperature (Tg) of specimens can be determined from the the DSC curves through the mid-point of transition. DSC curves through the mid-point of transition. 2.2.3. Dynamic Mechanics Analysis (DMA) 2.2.3. Dynamic Mechanics Analysis (DMA) A dynamic mechanical thermal analysis (DMA) instrument (NETZSCH, DMA242C, Bavaria, A dynamic mechanical thermal analysis (DMA) instrument (NETZSCH, DMA242C, Bavaria, Germany) used to toconduct conductthe theDMA DMAofofNBR NBR specimens. DMA performed under Germany) was was used specimens. TheThe DMA testtest waswas performed under the the uniaxial mode. Furthermore, Furthermore,the thedynamic dynamic tensile modulus measured a temperature uniaxial stretch stretch mode. tensile modulus waswas measured in ain temperature ◦ C to 30 ◦ C), under the rise rate of 3 ◦ C/min and the frequency of 1 Hz. sweep test (from − 60 sweep test (from −60 °C to 30 °C), under the rise rate of 3 °C/min and the frequency of 1 Hz. 2.2.4. and Wear WearTests Tests 2.2.4. Friction Friction and The experimentswere wereoperated operated using ultra-functional tribological testing machine The experiments using an an ultra-functional tribological testing machine under under a aball-on-disk ball-on-disk mode (CFT-I, Lanzhou, China) in ambient air (open air). Figure 2 shows the test mode (CFT-I, Lanzhou, China) in ambient air (open air). Figure 2 shows the test configuration In this this study, study, the configuration of of ball ball (steel)-on-plate (steel)-on-plate (rubber) (rubber) without without lubricants. lubricants. In the GCr15 GCr15 steel steel ball ball was set theascounterpart. The The surface roughness andand diameter of the steel ballball is about 0.80.8 µm and 5 mm, wasasset the counterpart. surface roughness diameter of the steel is about μm and respectively. Before the tribological test, the steel counterpart samples were cleaned in absolute 5 mm, respectively. Before the tribological test, the steel counterpart samples were cleaned in absoluteethyl alcohol, acetone and pure water withwith ultrasound for for 10 min, respectively. Finally, the ethyl alcohol, acetone and pure water ultrasound 10 min, respectively. Finally, thesamples sampleswere were dried in a stream of nitrogen. NBR specimenswere werevulcanized vulcanized into into square dried in a stream of nitrogen. TheThe NBR specimens squarewith withside sidelength length of of mm 50 mm and 2 mmthick. thick.Before Beforethe the tribological tribological test, specimens were bonded ontoonto the steel 50 and 2 mm test,the theNBR NBR specimens were bonded the steel disc for fixing. The tribological experiments were operated at the condition with a rotation speed of of disc for fixing. The tribological experiments were operated at the condition with a rotation speed 100 rpm rpm and and aa rotation applied constant normal loads were set up 1 N to 100 rotation radius radiusofof10 10mm. mm.The The applied constant normal loads were setfrom up from 1 N to ◦ 3.5 N. The friction tests were carried out at a room temperature of 25 °C with 25% relative humidity 3.5 N. The friction tests were carried out at a room temperature of 25 C with 25% relative humidity in in air. During sliding process, COF was measured continuously from frictional curve air. During thethe sliding process, thethe COF was measured continuously from thethe frictional curve bybystrain strain gauges and the wear rate (Rw) of rubber sample in volume can be obtained as follows: gauges and the wear rate (Rw ) of rubber sample in volume can be obtained as follows:
ΔV ∆V Rw =LN LN
Rw =
(2)
(2)
where ΔV is the wear volume loss of sample. L is the sliding distance during the frictional test, and where ∆Vnormal is the wear loss of wear sample. L isshown the sliding distance during frictional test,be and N N is the load. volume As the surface track in Figure 2b, the wearthe volume loss can isobtained the normal load. As (3). the surface wear track shown in Figure 2b, the wear volume loss can be obtained by Equation by Equation (3). Δ∆V V ==SS×× 2π2πR Rt t (3) (3)
where, is the the radius radiusof ofthe thefriction frictiontrack. track.S S the cross-section area of the friction track. where, R Rtt is is is the cross-section area of the friction track. Additionally, the counterpart counterpartsteel steelball ballininrubber/metal rubber/metal tribo-pairs be worn. In this Additionally, the tribo-pairs maymay alsoalso be worn. In this study, the wear behavior of the steel ball was related to a specific value k = r/R, where R is the radius study, the wear behavior of the steel ball was related to a specific value k = r/R, where R is the radius of the steel ball, and r is radius of the worn surface of the steel ball as shown in Figure 2c. of the steel and r is radius of the worn surface of the steel ball as shown in Figure 2c.
Figure The test testconfiguration configurationofofball-on-plate ball-on-plate tribometer: equipment diagram, (b) wear Figure 2. The tribometer: (a) (a) equipment diagram, (b) wear tracktrack of of NBR specimen specimen and NBR and(c) (c)wear wearofofcounterpart. counterpart.
Polymers Polymers 2018, 2018, 10, 10, 705 x FOR PEER REVIEW
6 of of 17 17
2.2.5. Scanning Electron Microscopy (SEM) 2.2.5. Scanning Electron Microscopy (SEM) High resolution images of the wear tracks on the surface of samples were observed by SEM (FEI High resolution images of the wear tracks on the surface of samples were observed by SEM Quanta200F, Hillsborough, OR, USA). Prior to the SEM measurement, all the NBR samples were (FEI Quanta200F, Hillsborough, OR, USA). Prior to the SEM measurement, all the NBR samples were coated with gold film by sputtering. coated with gold film by sputtering. In addition, to investigate the morphology and dispersibility of GF filler in the rubber bulk, the In addition, to investigate the morphology and dispersibility of GF filler in the rubber bulk, dumbbell-shaped specimens were tested using a low temperature impact test apparatus under liquid the dumbbell-shaped specimens were tested using a low temperature impact test apparatus under nitrogen environment, and the images of fracture surface were also observed using SEM. liquid nitrogen environment, and the images of fracture surface were also observed using SEM. 3. Results 3. Results 3.1. Characterization Characterization of of GF-Filled GF-Filled NBR NBR 3.1. Figure 33 shows results of of thethe filled andand unfilled nitrile butadiene rubber. Five Figure showsgroup grouptensile tensiletest test results filled unfilled nitrile butadiene rubber. group teststests under the the same parameters were carried out. Five group under same parameters were carried out.From Fromthe thestress-strain stress-straincurves curves of of different different NBR vulcanizates, the incorporation of glass flake within a rubber matrix does not have an obvious obvious NBR vulcanizates, the incorporation of glass flake within a rubber matrix does not have an change in the tensile behavior. The average tensile strength of different NBR specimens were listed change in the tensile behavior. The average tensile strength of different NBR specimens were listed in in Table 3. Compared with unfilled NBR, it can be observed that the average tensile strength Table 3. Compared with unfilled NBR, it can be observed that the average tensile strength increased increased flake composite. However, the tensile just increased with with glass with flake glass composite. However, the tensile strength juststrength increased slightly withslightly the filler sizethe of filler size of glass flake decreasing. glass flake decreasing.
sulphur-vulcanized NBR NBR specimens. specimens. Figure 3. Stress-strain curves of sulphur-vulcanized Table 3. 3. Mechanical Mechanical properties properties of of GF-filled GF-filled NBR NBR vulcanizates. vulcanizates. Table
Specimens Unfilled NBR GF-B GF-M GF-S Unfilled NBR GF-B GF-M GF-S Elastic modulus (MPa) 0.63 ± 0.01 0.63 ± 0.01 0.65 ± 0.01 0.66 ± 0.02 Elastic modulus (MPa) 0.63 ± 0.01 0.63 ± 0.01 0.65 ± 0.01 0.66 ± 0.02 Tensile strength (MPa) 6.24 ± 0.3 6.72 ± 0.3 7.07 ± 0.4 7.17 ± 0.3 Tensile strength (MPa) 6.24 ± 0.3 6.72 ± 0.3 7.07 ± 0.4 7.17 ± 0.3 Storage (MPa, modulus Storage modulus 20 ◦ C) 12.29 ± 0.03 13.28 ± 0.02 13.35 ± 0.02 13.51 12.29 ± 0.03 13.28 ± 0.02 13.35 ± 0.02 13.51 ± 0.02± 0.02 20 (A) °C) Shore (MPa, hardness 44 ± 0.8 47 ± 1 49 ± 0.8 53 ± 1 41.7 ± 1.1 43.8 ± 1 44.7 ± 1.2 46.6 ± 1.3 Fracture energy 10−3 , J/mm2 ) Shore(×hardness 44 ± 0.8 47 ± 1 49 ± 0.8 53 ± 1 (A) Fracture the energy Figure 4 displays heat flow as functions (− 70 to 20 ◦ C) by a DSC 41.7 ± 1.1 of the temperature 43.8 ± 1 44.7 ± 1.2 46.6 ± 1.3instrument. −3, J/mm2) (×10 From the DSC curves (shown in Figure 4), it can be noted that the temperature range which is related Specimens
to the mutation range of DSC baseline is consistent for each sample. Accordingly, the glass transition ◦ flow as functions of the temperature (−70 to 20 °C) by a DSC Figure 4(Tdisplays the heat temperatures g ) are around 0 C for all the NBR specimens as shown in Figure 4. It can be deduced instrument. From the DSC curves (shown in Figure 4), it caneffect be noted the temperature range that the introduction of GF filler would not show a significant on thethat movement of the polymer whichsegment. is related to the mutation range of DSC baseline is consistent for each sample. Accordingly, chain the glass transition temperatures (Tg) are around 0 °C for all the NBR specimens as shown in Figure
Polymers 2018, 10, x FOR PEER REVIEW
7 of 17
4. It can be deduced that the introduction of GF filler would not show a significant effect on the 7 of 17 polymer chain segment.
Polymers 2018, of 10,the 705 movement
Figure 4. Differential DifferentialScanning Scanning Calorimetry (DSC) curves of glass flake-filled and unfilled Calorimetry (DSC) curves of glass flake-filled and unfilled NBR NBR specimens. specimens.
It is tribological performances of materials could be influenced by the It is well wellknown knownthat thatthe the tribological performances of materials could be influenced bybasic the (a) mechanical performances. Thus, the hardness of different NBR specimens were measured by basic mechanical performances. Thus, the hardness of different NBR specimens were measured bya different NBR NBR asclerometer sclerometer(LX-A, (LX-A,Beijing, Beijing,China) China)according accordingtotothe theISO ISO 7619: 7619: 2004. 2004. The The hardness hardness of of different vulcanizatesare arelisted listedin inTable Table3.3.As Asshown shownininTable Table3,3,the theunfilled unfilled NBR behave a shore hardness vulcanizates NBR behave a shore hardness (A)(A) of of 44, approximately and the shore hardness after mixing with GF is larger than the unfilled NBR. 44, approximately and the shore hardness after mixing with GF is larger than the unfilled NBR. This is This is resulting from the of influence inorganic which could improve thecapacity load-carrying resulting from the influence inorganicoffillers which fillers could improve the load-carrying [28,29]. capacity [28,29]. Moreover, the hardness increased a little with the decreasing of GF size. It may be Moreover, the hardness increased a little with the decreasing of GF size. It may be corresponding with corresponding with the dispersibility of fillers in the polymer bulk and the bonding strength between the dispersibility of fillers in the polymer bulk and the bonding strength between interfaces of GF filler interfaces and rubber.of GF filler and rubber. In order ordertotoinvestigate investigate dispersibility of filler GF filler the rubber NBR surfaces fracture In thethe dispersibility of GF in thein rubber matrix,matrix, the NBRthe fracture surfaces were tested under a low temperature brittle fracture method. Figure 5a–c show the SEM were tested under a low temperature brittle fracture method. Figure 5a–c show the SEM images of images of specimens filled NBRfor specimens for and GF-B, GF-M and GF-S, respectively. From the cross-section filled NBR GF-B, GF-M GF-S, respectively. From the cross-section fracture surfaces, fracture surfaces, typically and tearingcan fracture features canthe beSEM found. Fromfor thethe SEM images, typically brittle and tearingbrittle fracture features be found. From images, most part, for the most part, the GF fillers are horizontal in the rubber structure after being cured by hydraulic the GF fillers are horizontal in the rubber structure after being cured by hydraulic vulcanizing press. vulcanizing press. it some can beGF found that GFthe fillers fell off on the fracture surfaces. Furthermore, it canFurthermore, be found that fillers fellsome off on fracture surfaces. Compared with Compared with three images as shown in Figure 5, it can be found that the bigger size GF is easier to three images as shown in Figure 5, it can be found that the bigger size GF is easier to fall off from fall off from the rubber mixture fracture surface. On the whole, the GF filler has a good uniform the rubber mixture fracture surface. On the whole, the GF filler has a good uniform dispersion in the dispersion in the rubber matrix. rubber matrix. Additionally, the DMA DMA investigation investigation of of NBR NBR vulcanizates vulcanizates is is observed observed using using aa DMA DMA instrument instrument Additionally, the (NETZSCH, DMA242C, DMA242C, Bavaria, Bavaria, Germany). DMA results results of of different different NBR NBR (NETZSCH, Germany). Figure Figure 66 shows shows the the DMA specimens. From the storage modulus (E’) curves as shown in Figure 6a, the storage modulus specimens. From the storage modulus (E’) curves as shown in Figure 6a, the storage modulus increased after after filled filled with with GF GF in in the the low-temperature low-temperature stage stage and and the the storage storage modulus modulus were were almost almost increased same for for the the three three NBR NBR specimens specimens with with different at 20 20 ◦°C, same different size size GF GF filler filler at C, which which was was about about 13.5 13.5 MPa. MPa. Moreover, the elasticity modulus after filled with GF showed no obvious change than the unfilled Moreover, the elasticity modulus after filled with GF showed no obvious change than the unfilled NBR NBR as in listed in3.Table 3. The loss modulus and(E”, factors Tan δ) were shown in respectively. Figure 6b,c, as listed Table The loss modulus and factors Tan δ)(E”, were shown in Figure 6b,c, respectively. With of theGF decrease GFreduced size, theasE”shown reduced as shown Figure 6b. Furthermore, With the decrease size, theofE” in Figure 6b. in Furthermore, Tan δ shows Tan the δ shows the characteristic of one peak which can be found in Figure 6c. All of the results indicate that characteristic of one peak which can be found in Figure 6c. All of the results indicate that the networks theall networks ofNBR all GF-filled samples are typically complete and continuous than a twoof GF-filled samplesNBR are typically complete and continuous rather thanrather a two-phase [30]. phase [30]. Additionally, the elasticity modulus of NBR is not changed obviously. What is more, Additionally, the elasticity modulus of NBR is not changed obviously. What is more, the rangethe of range of transition gets narrower the decreasing sizes of GFindicates which indicates that GF-S ashows transition gets narrower with the with decreasing sizes of GF which that GF-S shows highera higher miscible thanand GF-B and[31]. GF-M [31]. miscible than GF-B GF-M
Polymers 2018, 10, 705
8 of 17
Polymers 2018, 10, x FOR PEER REVIEW
8 of 17
Fracture Glass flake Fall off
(b)
(c)
Figure surfaces of of glass glass flake-filled flake-filled NBR: NBR: (a) (a) GF-B, GF-B, (b) (b) GF-M GF-M and and (c) (c) GF-S. GF-S. Figure 5. 5. SEM SEM images images of of fracture fracture surfaces
Polymers 2018, 10, 705 Polymers 2018, 10, x FOR PEER REVIEW
9 of 17 9 of 17
Figure 6. 6. The The Dynamic Dynamic Mechanics Mechanics Analysis Analysis (DMA) (DMA) patterns patterns of of NBR NBR vulcanizates: vulcanizates: (a) (a) storage storage modulus modulus Figure (E’), (b) (b) loss loss modulus modulus(E”) (E”)and and(c) (c)loss lossfactor factor(tan (tanδ). δ). (E’),
Polymers 2018, 10, 705
10 of 17
3.2. Tribological Properties of NBR Filled with GF Polymers 2018, 10, x FOR PEER REVIEW
10 of 17
The tribological behaviors of unfilled and GF-filled NBR samples under dry condition were tested Tribologicalfrictional Properties of NBR Filled with GF the schematic diagram is shown in Figure 2. In the using a 3.2. ball-on-disk instrument, which sliding test, The the different NBR samples with the electromotor at adry velocity of 100 rpm. tribological behaviors of rotate unfilledtogether and GF-filled NBR samples under condition were The coefficient of friction was measured the friction sensor. tested using a ball-on-disk frictional by instrument, which the schematic diagram is shown in Figure 2. Figure showstest, thethe typical curves COF with towith the rotating time under the applied In the7asliding different NBRof samples rotaterespect together the electromotor at a velocity of 100 rpm. The coefficient of friction was measured by the friction sensor. normal load of 3.5 N. From the COF curves, at the inception region, the COF rose sharply at first and Figure 7a shows the typical a curves of COFprocess with respect to sliding the rotating time Moreover, under the applied then dropped down. This represents running-in of the friction. after the normal load of 3.5 From the COF at running-in the inceptionprocess, region, the rose sharply at first region and steel ball sliding about 20N. min through thecurves, friction theCOF steady-state frictional then dropped down. This represents a running-in process of the sliding friction. Moreover, after the could be observed for each NBR sample. On the whole, the COF of NBR reduces after filled with GF. steel ball sliding about 20 min through the friction running-in process, the steady-state frictional Without the GF filler, the steady-state COF of NBR was fluctuating around 1.32. When with the GF region could be observed for each NBR sample. On the whole, the COF of NBR reduces after filled filler, the COF especially for GF-S NBR. Thewas COF of GF-S-filled decreased with GF.reduces Withoutobviously, the GF filler, the steady-state COF of NBR fluctuating aroundNBR 1.32. When with to about 1.08. Figure 7b shows the COF and the declining percent with respect to the normal loads from the GF filler, the COF reduces obviously, especially for GF-S NBR. The COF of GF-S-filled NBR 1 N to 3.5 N, after 1 h friction process. It can be seen that the COF decreases with the increasing of decreased to about 1.08. Figure 7b shows the COF and the declining percent with respect to thethe normalnormal load. The the points when friction variation loadsheat fromis1generated N to 3.5 N, on after 1 hcontacting friction process. It can be seen that occurs. the COF The decreases with of the increasing of the normal load. The heat is generated the contacting points whenbetween friction occurs. temperature further affects the COF. When the normal loadon increases, the temperature ball and The variation of temperature furthersoftens affectsthe thehardness COF. When the and normal load declines increases,[32,33]. the NBR enhances. The elevated temperature of NBR the COF temperature between ball and NBR enhances. The elevated temperature softens the hardness of NBR The extent of decline of COF is ranging from 1.4% to 5.8% for GF-B NBR at different normal loads. and the COF declines [32,33]. The extent of decline of COF is ranging from 1.4% to 5.8% for GF-B The GF-S NBR suffers the lowest COF in same experimental condition compared with other filled NBR. NBR at different normal loads. The GF-S NBR suffers the lowest COF in same experimental condition With the increasing of normal load (from 1 N to 3.5 N), the COF of GF-S NBR changes from 1.16 to 1.08. compared with other filled NBR. With the increasing of normal load (from 1 N to 3.5 N), the COF of When atGF-S a normal load offrom 1.5 N, the declines 21.1% compared NBR changes 1.16 toCOF 1.08. When at aabout normal load of 1.5 N, thewith COFunfilled declines NBR about sample. 21.1% Additionally, the wear behaviors of NBR samples are shown in Figure 8. Compared with the compared with unfilled NBR sample. unfilled NBR specimen, the resistanceofisNBR enhanced obviously with the filler of GF. From the line Additionally, the wear wear behaviors samples are shown in Figure 8. Compared with the unfilled NBR specimen, resistance is enhanced obviously withdecreased the filler ofas GF. From the line charts of wear rates as shownthe in wear Figure 8, it is found that the wear loss the friction time 2 /N to 0.34 2 /N charts wear rates as shown in Figure it is found the×wear decreased as the time increased. Forofunfilled NBR, the wear rates 8, change fromthat 0.63 10−9loss mm × friction 10−9 mm −9 mm2/N to 0.34 × 10−9 mm2/N with increased. For unfilled NBR, the wear rates change from 0.63 × 10 with the friction time increasing from 0.5 h to 3 h. The GF-S NBR also has the same trend, and its the drop friction time0.37 increasing h to to 30.05 h. The GF-S NBR alsoIthas the same of trend, and its wear 2 /N −9 mm 2 /N. wear rates from × 10−9from mm0.5 × 10 is because the contact region rates drop from 0.37 × 10−9 mm2/N to 0.05 × 10−9 mm2/N. It is because of the contact region enlarged enlarged with the increase in the friction time. Thus, the contact pressure declines due to the increasing with the increase in the friction time. Thus, the contact pressure declines due to the increasing of the of the area at contact interfaces. Furthermore, the wear rate is also affected by the glass flakes which area at contact interfaces. Furthermore, the wear rate is also affected by the glass flakes which appear appear on on the the sliding slidinginterfaces interfaces between steel specimens. 9 shows theloss wear between thethe steel ballball andand NBRNBR specimens. FigureFigure 9 shows the wear loss (wear ratio k) of thethe steel ball variednormal normalloads loadswith with different (wear ratio k) of steel ballcounterpart counterpart after after 33 h, h, under under varied different GFGF fillers. Compared with the NBRNBR condition, thethe wear showsa alittle little increase fillers. Compared withunfilled the unfilled condition, wearratio ratioof ofsteel steel ball ball shows increase of GF. Furthermore, the wear of steel only increases slightlywith withthe theincrease increase in with thewith fillerthe offiller GF. Furthermore, the wear ratioratio of steel ballball only increases slightly in the normal and the influence of filler sizewear on the wear loss of steel ball counterpart is not the normal load, and load, the influence of filler size on the loss of steel ball counterpart is not obvious. obvious. It is slightly with the of GF filler size. It is slightly decreased withdecreased the reduction of reduction GF filler size.
(a)
Figure 7. Cont.
Polymers 2018, 10, x FOR PEER REVIEW Polymers 2018, 10, x FOR PEER REVIEW Polymers 2018, 10, 705 Polymers 2018, 10, x FOR PEER REVIEW
11 of 17 11 of 17 11 of 17 11 of 17
(b) (b) (b)
Figure 7. (a)7.Coefficients of friction (COF) timefor forfilled filled and unfilled NBR specimens Figure (a) Coefficients of friction (COF)versus versus friction friction time and unfilled NBR specimens Figure 7. (a) (a) Coefficients Coefficients of friction friction (COF) versus friction time for for filled filled and unfilled NBR specimens Figure 7. of (COF) versus friction time and unfilled NBR specimens a rotation speed of 100 rpm andnormal normal load load of 3.5 thethe declining percent of of underunder a rotation speed of 100 rpm and 3.5 N; N;(b) (b)COF COFand and declining percent under a rotation speed of 100 rpm and normal load of 3.5 N; (b) COF and the declining percent of underNBR a rotation speed of 100 rpm and normal load of 3.5 N; (b) COF and the declining percent of NBR specimens under different normal loads. NBR specimens under different normal loads. NBR specimens under different normal loads. specimens under different normal loads.
Figure 8. The wear ratios versus sliding time for filled and unfilled NBR specimens.
Figure 8. 8. The wear ratios versus sliding time time for filled filled and unfilled unfilled NBR specimens. specimens. Figure Figure 8. The wear ratios versus sliding sliding time for for filled and and unfilled NBR NBR specimens.
Figure 9. Wear ratios of counterpart steel ball against unfilled and GF-filled NBR specimens under normal load 1.5 N, 2.5 N and 3.5 N.
Figure Figure 9. 9. Wear Wear ratios ratios of of counterpart counterpart steel steel ball ball against against unfilled unfilled and and GF-filled GF-filled NBR NBR specimens specimens under under Figure 9. Wear ratios of counterpart steel ball against unfilled and GF-filled NBR specimens under normal normal load load 1.5 1.5 N, N, 2.5 2.5 N N and and 3.5 3.5 N. N. normal load 1.5 N, 2.5 N and 3.5 N.
Polymers 2018, 10, 705 Polymers 2018, 10, x FOR PEER REVIEW
12 of 17 12 of 17
Figure surface wear tracks of different NBR NBR specimens. For theFor unfilled NBR sample, Figure 10 10shows showsthe the surface wear tracks of different specimens. the unfilled NBR the typical wear patterns for rubber materials are observed as shown in Figure 10a. From the abraded sample, the typical wear patterns for rubber materials are observed as shown in Figure 10a. From the surface unfilledofNBR, a series are ridges generated on the wear which are almost abradedofsurface unfilled NBR,ofaparallel series ofridges parallel are generated onsurface, the wear surface, which perpendicular to the sliding previousAs studies indicated, periodicthe wear patterns of are almost perpendicular to direction. the slidingAs direction. previous studiesthe indicated, periodic wear surface are a common feature in the viscoelastic materials [34,35]. The formation of the periodic wear patterns of surface are a common feature in the viscoelastic materials [34,35]. The formation of the patterns (wavypatterns pattern)(wavy is duepattern) to a complex compression and extension strain cycle in the cycle contact periodic wear is due to a complex compression and extension strain in interfaces, and this is a typical wear phenomenon for special rubber-like materials [36–39]. Moreover, the contact interfaces, and this is a typical wear phenomenon for special rubber-like materials [36– some curled debris can be seen on the surface as shown in Figure 10a. It can be deduced 39]. Moreover, some curled debris canworn be seen on the worn surface as shown in Figure 10a. Itthat canthe be local adhesion waslocal present between thepresent rubberbetween and steelthe surface, and then thesurface, adhesive rubber deduced that the adhesion was rubber and steel and thenwas the lengthened whichwas is repeated in the friction with process. the particular appearance for adhesive rubber lengthened which is process. repeatedCompared in the friction Compared with the the unfilled NBR, littlefor pitsthe and small ridges occurpits in the of GF-filled NBR surfaces specimens particular appearance unfilled NBR, little andworn smallsurfaces ridges occur in the worn of shown as NBR Figure 10b–d. Inshown summary, the COF of NBR filled withthe GFCOF decreased and its anti-wear GF-filled specimens as Figure 10b–d. In summary, of NBR filled with GF performance Figure 11a shows the worn surface of the steel morphology ball against decreased andenhanced. its anti-wear performance enhanced. Figure 11amorphology shows the worn surface unfilled NBR sample. can be NBR foundsample. that noItobvious damage on the steel appeared ball counter of the steel ball againstItunfilled can be found thatappeared no obvious damage on surface. Moreover, Figure 11b shows the damage morphology of the steel ball against the GF-S-filled the steel ball counter surface. Moreover, Figure 11b shows the damage morphology of the steel ball NBR sample. ComparedNBR withsample. Figure 11a, serious with ploughs were observed the steel ballobserved surface due against the GF-S-filled Compared Figure 11a, seriouson ploughs were on to GF ball particles present onthe theGF contact interfaces. Therefore, the main wear mechanism themain steel thethe steel surface due to particles present on the contact interfaces. Therefore,ofthe ball the abrasive wear bythe theabrasive hard glass particles the rubber, which could wearwas mechanism of the steelcaused ball was wear causedembedded by the hardinglass particles embedded accelerate the wear metal.accelerate the wear of metal. in the rubber, whichofcould
(b)
(a)
Sliding direction of rubber
Sliding direction of rubber
(c)
(d)
Sliding direction of rubber
Sliding direction of rubber
Figure 10. 10. SEM SEMimages imagesofofabrasive abrasivesurfaces surfaces unfilled NBR, GF-B-, (c) GF-M-, andGF-S-filled (d) GF-SFigure of of (a)(a) unfilled NBR, (b) (b) GF-B-, (c) GF-M-, and (d) filled NBR specimens. NBR specimens.
Polymers 2018, 10, 705
13 of 17
Polymers 2018, 10, x FOR PEER REVIEW
13 of 17
(a)
(b)
Polymers 2018, 10, x FOR PEER REVIEW
13 of 17
(a)
(b)
Figure 11. 11. Worn Worn surface of steel ball against (a) unfilled NBR and (b) GF-S-filled NBR specimens. Figure
4. Discussions Figure 11. Worn surface of steel ball against (a) unfilled NBR and (b) GF-S-filled NBR specimens. 4. Discussions 4.1. Effect of GF Filler on the Friction of NBR
The results of tribological tests indicated that the COF of NBR reduced with filler of micro glass 4.1. Effect of GF Filler on the Friction of NBR because the contact contact surface surface changed changed with with GF GF filler. filler. Figure Figure 12 shows shows the the schematic schematic view flakes. This is because The results of tribological tests indicated that the COF of NBR reduced with filler of micro glass frictional NBR against steel ball. On one hand, discovered of contact and process of On one hand, it can be flakes. This is because the contact surface changed with GF filler. Figure 12 shows the schematic viewfrom Figure that the contact area and deformation of NBR decrease with GF-filled NBR. This is This caused 12a,b that the contact area and deformation of NBR with NBR. is of12a,b contact and frictional process of NBR against steel ball. On decrease one hand, it can GF-filled be discovered from by theFigure enhancement hardness and load-carrying capacity with GFwith filler. the On other GF caused by the enhancement of hardness load-carrying capacity GFOn filler. thehand, other 12a,b thatofthe contact area andand deformation of NBR decrease with GF-filled NBR. This the ishand, appeared atby the during and the sliding Accordingly, theOn real range at caused thefriction enhancement of hardness load-carrying capacity with GF filler. thecontact other hand, the GF appeared at the interface friction interface during theprocess. sliding process. Accordingly, the real contact the GF appeared at the friction interface during the sliding process. Accordingly, the real contact the interfaces between between the NBR the sample steeland ball steel decreased due to thedue GFto filler. regard to range at the interfaces NBRand sample ball decreased the With GF filler. With at theand interfaces between thenormal NBR sample andpressure, steel ballsteel decreased due tosurface the GFtangential filler. Withforce, GF-Brange and GF-M, due to the contact pressure, rough ball surface and regard to GF-B GF-M, due to the contact normal rough steel ball and tangential regardGF to GF-B GF-M, due the contact normal pressure, rough steel ball as surface andintangential the bigger becomes warped uptoand could offfall from rubber surface shown Figure 10b–d. force, the bigger GFand becomes warped up and fall could offthe from the rubber surface as shown in Figure force, the bigger GF becomes warped up and could fall off from the rubber surface as shown in Figure DuringDuring the sliding process,process, the chips GF appear the rubber This phenomenon leads to 10b–d. the sliding theofchips of GF on appear on thesurface. rubber surface. This phenomenon 10b–d. During the sliding process, the chips of GF appear on the rubber surface. This phenomenon the reduction of the contact region between the NBR sample and steel ball. Thus, the adhesive force leads to the reduction of the contact region between the NBR sample and steel ball. Thus, the adhesive leads to the reduction of the contact region between the NBR sample and steel ball. Thus, the adhesive between the rubber and steel ball interfaces reduces with GF filler. As a result, the COF decreases after force force between the the rubber and steel reduceswith withGFGFfiller. filler. a result, the COF between rubber and steelball ballinterfaces interfaces reduces As As a result, the COF filleddecreases with after GF. after Moreover, theGF. GF-S with the the smaller is not easily warped compared with decreases filled with Moreover, with thesmaller smaller size isupnot easily warped up filled with GF. Moreover, the GF-S GF-Ssize with the size is not easily warped upGF-B and GF-M in this study, which can be found in Figures 10d and 12d. Thus, the NBR sample filled with compared with GF-B and GF-M in this study, which can be found in Figures 10d and 12d. Thus, the compared with GF-B and GF-M in this study, which can be found in Figures 10d and 12d. Thus, the GF-S NBR shows afilled lower COF comparing GF-BCOF and GF-M samples. sample filled with GF-S shows lower COF comparing totoGF-B and GF-M samples. NBR sample with GF-S shows atoalower comparing GF-B and GF-M samples. (a) (a) (c)
(c)
Sliding direction
Sliding direction (b)
(b)
(d)
(d)
Sliding direction
Figure 12. 12. Schematic diagram ofofcontact andwear wearofofdifferent different NBR specimens. Sliding direction Figure Schematic diagram contact and NBR specimens.
Figure 12. Schematic diagram of contact and wear of different NBR specimens.
Polymers 2018, 10, 705
14 of 17
4.2. Effect of GF Filler on the Wear Behavior of NBR As mentioned above, Figure 8 displays the relationship of the wear rate with respect to the sliding time. It can be seen that Rw is dependent on the sliding time. From Figure 8, we can easily observe that the Rw decreases as the increase of time. It is because of the augment of contact region and the effect of rolling of debris on the contact interfaces during the successive wear process. As shown in Figure 10a, due to the repeated friction and contact process, the crack might appear when the elongation reaches a critical value, and the wear debris are crimping and rolling by the edge of contact during the consecutive friction process [35]. Afterwards, the detached rubber particles with the form of curled debris appear from the surface layer, when the crack is propagating to meet another. Furthermore, this outcome contributes to the production of curl pattern on the rubber surface and enhances morphology matching at the contact interfaces between the NBR and steel ball. Once curl-shaped worn pattern occurs, it will expand gradually under the condition of reciprocating pressure and drag during the sliding process. Finally, the curl-shaped debris are torn apart from the rubber matrix. Many researchers studied the relationship of wear morphology and the extension of crack [36]. The fatigue-fracture mechanism and decomposition caused by mechanochemistry could be used to explain the wear failure of rubber [35,40]. Gent and Pulford [41] noted that the occurrence of the oil grinding dust marks the occurrence of mechanochemistry decomposition. In this study, the transfer film is not observed, and the debris were dry. Therefore, the formation of curled debris is due to the adhesive wear and fatigue wear. This indicated that the wear mechanism of NBR is mainly adhesive and fatigue-fracture failure. The original NBR shows a higher wear rate at the initial of sliding process, and then decreases with the increase in friction time. When the time reaches to 2 h, the wear rate tends to be stable as shown in Figure 8. Compared with the original NBR, the three GF-filled NBR specimens behave a better wear-resistant. Additionally, the wear loss shows a slight reduce with the decrease of GF size. This is ascribed to the reinforcement of the micro glass flakes. The GF-filled NBR samples behave a higher hardness compared to that of the unfilled sample which is indicated in Table 3. The wear surface patterns are different between the filled and unfilled NBR specimens, as shown in Figure 10. The degree of surface wear of GF-filled NBR became lower than that for the unfilled NBR. The typical curl-shaped patterns on the wear surface were observed as shown in the SEM images. For the GF-filled NBR specimens, the GF chips were obviously noticed on the wear surfaces of GF-B-, GF-M- and GF-S-filled NBR as shown in Figure 10b–d. From the schematic diagram (Figure 12c,d), it is indicated that the GF chips could improve the contact interface, which can prevent the extension of rubber curl-shaped patterns and reduce the crack at the surface due to the decrease of the stress concentration. Thus, the wear-resistant of NBR enhanced after mixed with GF fillers. Additionally, the fracture energy can describe the energy change of extension of crack, which is a common application for analysis of the fatigue behaviors of polymers [42]. The average fracture energy is listed in Table 3. For each specimen, five tests are applied under the same condition. From Figure 13, it can been seen that the value changes of GF-B-, GF-M- and GF-S-filled NBR are slight. Generally, previous researches [43–45] indicated that the polymer fracture energy can be influenced by properties of filler, such as dispersity, volume fraction, morphology, interfacial strength, etc. In this study, the fillers’ morphologies and volume fractions are basically the same which can be found in Figure 1 and Table 1. Furthermore, from the fracture surfaces investigated by a low temperature brittle fracture method as shown in Figure 5, most of the GF fillers are horizontal in the rubber matrix after being cured by the hydraulic vulcanizing press process and the filler has a good dispersity for three GF fillers. Therefore, the fracture energy is majorly affected by the interfacial strength.
Polymers 2018, 10, 705 Polymers 2018, 10, x FOR PEER REVIEW
15 of 17 15 of 17
Figure 13. 13. Fracture energy and and semiempirical semiempirical constant B of of interfacial interfacial strength strength between between glass flake flake and rubber matrix. rubber matrix.
As previous previous study study indicated, indicated, the the Pukánszky Pukánszky equation equation is is generally generally recognized recognized to to the the filled filled As elastomer materials [46]. The tensile strength of elastomer materials and the relationship of volume elastomer materials [46]. The tensile strength of elastomer materials and the relationship of volume fraction to to interfacial interfacial strength strength can can be be shown shown as as follows: follows: fraction
11−−ννf f σσm exp( exp( Bν = 1 + 2.5ν σσcc = Bνf )f ) 1 + 2.5ν f m f !
(4) (4)
where σc and σm are the tensile strength of filled and unfilled polymer. B is semiempirical constant, where σc and σm are the tensile strength of filled and unfilled polymer. B is semiempirical constant, and the higher value of B shows stronger interfacial strength at the interface of the filler and polymer. and the higher value of B shows stronger interfacial strength at the interface of the filler and polymer. vf is the filler’s volume fraction. The semiempirical constant B of GF-filled NBR of GF-B, GF-M and vf is the filler’s volume fraction. The semiempirical constant B of GF-filled NBR of GF-B, GF-M and GF-S are about 4.72, 5.64 and 5.90, respectively. As Equation (3) indicated, the interfacial strength GF-S are about 4.72, 5.64 and 5.90, respectively. As Equation (3) indicated, the interfacial strength between GF filler and rubber matrix behaves the same change in the fracture energy. Furthermore, between GF filler and rubber matrix behaves the same change in the fracture energy. Furthermore, the specific surface area increased with the decreasing of GF size. This leads to the enhancement of the specific surface area increased with the decreasing of GF size. This leads to the enhancement of the interfacial strength at interface of the GF filler and rubber matrix. This is also contributing to the the interfacial strength at interface of the GF filler and rubber matrix. This is also contributing to the improvement of the interfacial strength and is beneficial for enhancing the wear resistance of NBR. improvement of the interfacial strength and is beneficial for enhancing the wear resistance of NBR. 5. Conclusions 5. Conclusions The nitrile rubbers filled with micro glass flake were prepared with different size fillers. Universal The nitrile filled with glass flake prepared with different size test machine, DSCrubbers and DMA were usedmicro to characterize thewere NBR specimens and the frictional andfillers. wear Universal test machine, DSC and DMA were used to characterize the NBR specimens and the properties were investigated using a ball-on-flat mode tribometer. frictional and wear properties were investigated using a ball-on-flat mode tribometer. The tribological test results showed that the COF reduced and the wear resistance improved for testglass results showed that the COF reduced the could wear resistance for NBR The filledtribological with GF. The flake fillers appeared on the worn and surface restrain theimproved present and NBR filledofwith GF. The glass flake fillers appeared the improve worn surface could restrain property the present extension the curled patterns on the surface, whichon could the wear resistance of and extension of the curled patterns on the surface, which could improve the wear resistance rubber. However, the hard glass flake fall off from the rubber matrix would accelerate the wear of property of metal rubber. However, the hard glass flake fall off from themanifested rubber matrix would counterpart ball. Furthermore, the analysis of fracture surface that the filleraccelerate of micro the wear of counterpart metal ball. Furthermore, the analysis of fracture surface manifested that the glass flake possesses a good dispersity in the rubber matrix and good interfacial action. Moreover, filler of micro glass flake possesses a good dispersity in the rubber matrix and good interfacial action. filler with smaller size is more conducive to enhance the interfacial strength of the polymer matrix. Moreover, filler with size is more to enhance theisinterfacial of the the polymer This contributes to thesmaller improvement of theconducive surface interaction and beneficial strength for enhancing wear matrix. This contributes to the improvement of the surface interaction and is beneficial for enhancing resistance of NBR. the wear resistance of NBR. Author Contributions: Y.G. and Z.C. designed the experiments; H.T. and Z.C. performed the experiment; Author Contributions: Y.G. and Z.C.the designed the experiments; H.T. andthe Z.C. performed the experiment; Y.G., H.T., Z.C. and D.W. analyzed data; D.W. and S.Z. proposed idea of the project; Y.G. andY.G., H.T. H.T., Z.C. and D.W. analyzed the data; D.W. and S.Z. proposed the idea of the project; Y.G. and H.T. wrote and wrote and Z.C. finalized the manuscript; all authors have read and approved the final version of this manuscript. Z. C. finalized the manuscript; all authors have read and approved the final version of this manuscript.
Polymers 2018, 10, 705
16 of 17
Funding: This research is funded by the National Natural Science Foundation of China (No. 51375495), the Tribology Science Fund of State Key Laboratory of Tribology, China Scholarship Council, and the Science Foundation of China University of Petroleum, Beijing (No. 2462017BJB06, C201602). Conflicts of Interest: The authors declare no conflict of interest.
References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15.
16.
17.
18. 19. 20. 21. 22.
Zhang, S.W. Tribology of Elastomers; Elsevier: Amsterdam, The Netherlands, 2004. Choi, S.S.; Ha, S.H. Water swelling behaviors of silica-reinforced NBR composites in deionized water and salt solution. J. Ind. Eng. Chem. 2010, 16, 238–242. [CrossRef] Li, Y.; Wang, Q.; Wang, T.; Pan, G. Preparation and tribological properties of grapheme oxide/nitrile rubber nanocomposites. J. Mater. Sci. 2012, 47, 730–738. [CrossRef] Zhang, S.W.; Liu, H.C.; He, R.Y. Mechanisms of wear of steel by natural rubber in water medium. Wear 2004, 256, 226–232. [CrossRef] Briscoe, B.J.; Sinha, S.K. Wear of Polymers. Proceed. Inst. Mech. Eng. Part J J. Eng. Tribol. 2002, 216, 401–413. [CrossRef] Fedorko, G.; Molnár, V.; Živˇcák, J.; Husáková, N. Failure analysis of textile rubber conveyor belt damaged by dynamic wear. Eng. Fail. Anal. 2013, 28, 103–114. [CrossRef] Lv, X.; Wang, S.; Huo, X. The effect of water content in crude oil on the tribological behavior of screw pump’s stator rubber. Chin. Sci. Bull. 2014, 59, 2868–2874. [CrossRef] Cao, Z.; Wang, D.; Cong, C.; Wang, Y.; Zhou, Q. Dependence of abrasion behavior on cross-linked heterogeneity in unfilled nitrile rubber. Tribol. Int. 2014, 69, 141–149. [CrossRef] Chen, B.; Wang, J.; Yan, F. Friction and wear behaviors of several polymers sliding against GCr15 and 316 steel under the lubrication of sea water. Tribol. Lett. 2011, 42, 17–25. [CrossRef] Shen, M.X.; Dong, F.; Zhang, Z.X.; Meng, X.K.; Peng, X.D. Effect of abrasive size on friction and wear characteristics of nitrile butadiene rubber (NBR) in two-body abrasion. Tribol. Int. 2016, 103, 1–11. [CrossRef] Guo, Y.; Wang, J.; Li, K.; Ding, X. Tribological properties and morphology of bimodal elastomeric nitrile butadiene rubber networks. Mater. Des. 2013, 52, 861–869. [CrossRef] Friedrich, K.; Zhang, Z.; Schlarb, A.K. Effects of various fillers on the sliding wear of polymer composites. Compos. Sci. Technol. 2005, 65, 2329–2343. [CrossRef] Zhang, S.W. Recent developments of green tribology. Surf. Topogr. Metrol. Prop. 2016, 4, 023004. [CrossRef] Schadler, L.S. Polymer-Based and Polymer-Filled Nanocomposites; Wiley-VCH Verlag GmbH & Co. KGaA: Weinheim, Germany, 2003. Gatos, K.G.; Karger-Kocsis, J. Effect of the aspect ratio of silicate platelets on the mechanical and barrier properties of hydrogenated acrylonitrile butadiene rubber (HNBR)/layered silicate nanocomposites. Eur. Polym. J. 2007, 43, 1097–1104. [CrossRef] Malas, A.; Das, C.K.; Das, A.; Heinrich, G. Development of expanded graphite filled natural rubber vulcanizates in presence and absence of carbon black: Mechanical, thermal and morphological properties. Mater. Des. 2012, 39, 410–417. [CrossRef] Laura, D.M.; Keskkula, H.; Barlow, J.W.; Paul, D.R. Effect of rubber particle size and rubber type on the mechanical properties of glass fiber reinforced, rubber-toughened nylon 6. Polymer 2003, 44, 3347–3361. [CrossRef] Rezende, C.A.; Bragança, F.C.; Doi, R.T.; Lee, L.T.; Galembeck, F.; Boué, F. Natural rubber-clay nanocomposites: Mechanical and structural properties. Polymer 2010, 51, 3644–3652. [CrossRef] Tian, M.; Liang, W.; Rao, G.; Zhang, L.; Guo, C. Surface modification of fibrillar silicate and its reinforcing mechanism on FS/rubber composites. Compos. Sci. Technol. 2005, 65, 1129–1138. [CrossRef] Guo, Y.; Cao, Z.; Wang, D.; Liu, S. Improving the friction and abrasion properties of nitrile rubber hybrid with hollow glass beads. Tribol. Inter. 2016, 101, 122–130. [CrossRef] Tian, W.; Liu, L.; Meng, F.; Liu, Y.; Li, Y.; Wang, F. The failure behavior of an epoxy glass flake coating/steel system under marine alternating hydrostatic pressure. Corros. Sci. 2014, 86, 81–92. [CrossRef] Sathiyanarayanan, S.; Azim, S.S.; Venkatachari, G. Corrosion protection coating containing polyaniline glass flake composite for steel. Electrochem. Acta 2008, 53, 2087–2094. [CrossRef]
Polymers 2018, 10, 705
23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46.
17 of 17
Wang, G.; Yang, J. Influences of glass flakes on fire protection and water resistance of waterborne intumescent fire resistive coating for steel structure. Prog. Org. Coat. 2011, 70, 150–156. [CrossRef] Franklin, P.; Wood, D.J.; Bubb, N.L. Reinforcement of poly(methyl methacrylate) denture base with glass flake. Dent. Mater. 2005, 21, 365–370. [CrossRef] [PubMed] Vu-Khanh, T.; Fisa, B. Impact fracture of glass flake reinforced polypropylene. Polym. Compos. 1986, 7, 375–382. [CrossRef] Klaas, N.V.; Marcus, K.; Kellock, C. The tribological behavior of glass filled polytetrafluoroethylene. Tribol. Int. 2005, 38, 824–833. [CrossRef] Cho, K.; Lee, D. Effect of molecular weight between cross-links on the abrasion behavior of rubber by a blade abrader. Polymer 2000, 41, 133–140. [CrossRef] Chang, L.; Zhang, Z.; Ye, L.; Friedrich, K. Synergistic effects of nanoparticles and traditional tribo-fillers on sliding wear of polymeric hybrid composites. Tribol. Interf. Eng. Ser. 2008, 55, 35–61. Guo, Y.; Wang, D.; Zhang, S. Tribological behavior of in situ Ag nanoparticles/polyelectrolyte composite molecular deposition films. Appl. Surf. Sci. 2010, 256, 1714–1719. [CrossRef] Von Lockette, P.R. Examination of the effects of computationally determined network topology on an analytical constitutive model for bimodal elastomers. Polymer 2008, 49, 5158–5168. [CrossRef] Varughese, K.T.; Nando, G.B.; De, P.P.; Sanyal, S.K. Miscible blends from rigid poly(vinyl chloride) and epoxidized natural rubber. J. Mater. Sci. 1988, 23, 3903–3909. [CrossRef] Ashby, M.F.; Abulawi, J.; Kong, H.S. Temperature maps for frictional heating in dry sliding. Tribol. Trans. 1991, 34, 577–587. [CrossRef] Lan, P.; Zhang, Y.; Dai, W.; Andreas, A.P. A phenomenological elevated temperature friction model for viscoelastic polymer coatings based on nanoindentation. Tribol. Int. 2018, 119, 299–307. [CrossRef] Aharoni, S.M. Wear of polymer by roll-formation. Wear 1973, 25, 309–327. [CrossRef] Fukahori, Y.; Yamazaki, H. Mechanism of rubber abrasion. Part I: Abrasion pattern formation in natural rubber vulcanizate. Wear 1994, 171, 195–202. [CrossRef] Barquins, M. Adherence, friction and wear of rubber-like materials. Wear 1992, 158, 87–117. [CrossRef] Gatos, K.G.; Kameo, K.; Karger-Kocsis, J. On the friction and sliding wear of rubber/layered silicate nanocomposites. Express Polym. Lett. 2007, 1, 27–31. [CrossRef] Felhös, D.; Karger-Kocsis, J. Tribological testing of peroxide-cured EPDM rubbers with different carbon black contents under dry sliding conditions against steel. Tribol. Int. 2008, 41, 404–415. [CrossRef] Karger-Kocsis, J.; Felhos, D.; Xu, D.; Schlarb, A.K. Unlubri-cated sliding and rolling wear of thermoplastic dynamic vulcanizates (Santoprene® ) against steel. Wear 2008, 265, 292–300. [CrossRef] Fukahori, Y.; Yamazaki, H. Mechanism of rubber abrasion Part 3: How is friction linked to fracture in rubber abrasion? Wear 1995, 188, 19–26. [CrossRef] Gent, A.N.; Pulford, C.T.R. Mechanisms of rubber abrasion. J. Appl. Polym. Sci. 1983, 28, 943–960. [CrossRef] Reis, J.M.L.; Ferreira, A.J.M. A contribution to the study of the fracture energy of polymer concrete and fibre reinforced polymer concrete. Polym. Test. 2004, 23, 437–440. [CrossRef] Miwa, M.; Takeno, A.; Hara, K.; Watanabe, A. Volume fraction and temperature dependence of mechanical properties of silicone rubber particulate/epoxy blends. Compos. Part A 1995, 26, 371–377. [CrossRef] Okada, O.; Keskkula, H.; Paul, D.R. Fracture toughness of nylon 6 blends with maleated ethylene/propylene rubbers. Polymer 2000, 41, 8061–8074. [CrossRef] Ferreira, J.A.M.; Capela, C.; Costa, J.D. A study of the mechanical behavioue on fibre reinforced hollow microspheres hybrid composites. Compos. Part A 2010, 41, 345–352. [CrossRef] Pukánszky, B. Influence of interface interaction on the ultimate tensile properties of polymer composites. Compos. Part A 1990, 21, 255–262. [CrossRef] © 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
