- Email: [email protected]
h-BN composites
Journal of Alloys and Compounds 723 (2017) 345e353
Contents lists available at ScienceDirect
Journal of Alloys and Compounds journal homepage: http://www.elsevier.com/locate/jalcom
Surface modification of h-BN and its influence on the mechanical properties of CuSn10/h-BN composites Ting Li a, b, Danqing Yi a, b, Jun Hu a, b, Jiao Xu a, b, Junlei Liu a, b, Bin Wang a, b, * a b
School of Material Science and Engineering, Central South University, Changsha, Hunan 410083, PR China Light Alloy Research Institute, Central South University, Changsha, Hunan 410083, PR China
a r t i c l e i n f o
a b s t r a c t
Article history: Received 14 April 2017 Received in revised form 23 June 2017 Accepted 24 June 2017 Available online 27 June 2017
CuSn10 matrix composites with hexagonal boron nitride (h-BN) in concentrations of 4, 7, 10, and 14 vol.% were prepared using a powder metallurgy process. The effect of modifying the h-BN with anionic polyacrylamide (APAM) by high-energy ball milling was investigated. The bonding mechanism between h-BN and APAM was characterized by Fourier transform infrared spectroscopy (FT-IR), X-ray photoelectron spectroscopy (XPS), and X-ray diffraction (XRD). The microstructure, hardness, tensile strength, and bending strength of sintered composites were investigated. The composite CuSn10-4 vol%h-BN showed the lowest porosity of 20.1%. The long APAM chains could be introduced on the h-BN surface without altering its crystal structure. The surface modified h-BN exhibited less agglomeration in the matrix than pristine h-BN, due to the physical entanglement and chemical bonding between h-BN and APAM. Comparing the composites with 4 vol.% h-BN, without and with APAM modification, a reduction in the porosity from 20.1% to 5.5%, and increase in the tensile strength from 134 to 147 MPa, and a significantly increased in the bending strength from 238 to 351 MPa, respectively, were observed. © 2017 Published by Elsevier B.V.
Keywords: h-BN Anionic polyacrylamide Surface modification Copper matrix composite Powder metallurgy
1. Introduction Many tribological system working in high-temperature aggressive environments demand combinations of a low friction coefficient, good high-temperature oxidation resistance, and good wear resistance [1]. Tin bronze matrix powder metallurgy (PM) composites with excellent tribological properties and good heat conductivity and corrosion resistance have been widely investigated [2e4]. Traditional solid lubricants, such as graphite and MoS2, are good choices for tin bronze friction materials, where their lamellar structures are beneficial for decreasing friction, but most such lubricants can only be used under ambient temperature [3,5e7]. Hexagonal boron nitride (h-BN), is an inorganic analogue of graphene and has a lamellar structure consisting of a stack of hexagonal sheets with a strong covalent bond between boron and nitrogen atoms. The sheets are held together by weak van der Waals forces that allow shearing when force is applied parallel to the sheets. Thus, it provides the expected friction reduction and
* Corresponding author. School of Material Science and Engineering, Central South University, Changsha, Hunan 410083, PR China. E-mail address: [email protected] (B. Wang). http://dx.doi.org/10.1016/j.jallcom.2017.06.264 0925-8388/© 2017 Published by Elsevier B.V.
results in very efficient lubrication over a wide temperature range [1,8e13]. However, due to the surface incompatibility between hBN and the matrix alloy, the physical mixing of h-BN particles leads to phase segregation between the h-BN and matrix, giving rise to agglomeration of h-BN particles, which may negatively affect the physical and mechanical properties of the resulting composites [1,14e16]. Thus, reducing the agglomeration of h-BN particles and improving the interfacial adhesion between h-BN and the matrix surface have been the most important issues for the development of a tin bronze/h-BN composite. To break down the particle agglomerates and improve the dispersibility of the particles in the matrix, surface modification is an effective and commonly used method [17]. Silicon carbide has previously been modified by a silane coupling agent and hexadecyl iodiele [18]. Silane coupling agents with different carbon chains (C3 and C16) were introduced on the BN surface to improve the affinity of BN for the epoxy resin [19]. In the work of Li et al., TiO2 was modified by KH550 through a silanization reaction [20]. Anionic polyacrylamide (APAM) is produced during the polymerization reaction of acrylamide and has both amide and carboxyl functional groups [21]. APAM gel is a typical soft material that has a unique, three-dimensionally cross-linked network structure swollen with a large amount of water. The unique structure of these
346
T. Li et al. / Journal of Alloys and Compounds 723 (2017) 345e353
systems means they can be safely applied in materials processing and hence the study of these materials is of high utility. Currently, APAM is mainly applied in cosmetics, pharmaceutics and paint production, minerals flotation, oil recovery, agriculture, and environmental protection applications [22e24]. In order to realize homogeneous dispersion of h-BN in a tin bronze matrix, here we propose to prepare CuSn10/h-BN composites using APAM as a surface modifier. The modification mechanism of APAM on the surface of h-BN was studied. In addition, the influence of the APAM on the mechanical properties of the composite made with the modified h-BN was investigated. 2. Experimental 2.1. Materials and processing CuSn10 powder with an average particle size ~35 mm with a purity of 99.99% was used as the metal matrix material. Commercial h-BN (99.9% purity) was used as a lubrication material while the other powders (Ni, Fe, Mn, and WC) were used as reinforcement materials. All these powders were passed through a 200 mesh sieve. APAM has the chemical formula eCH2eCH(CONH2) eCH2eCH(COONa)e and an average molecular weight of 14000000. Fig. 1 shows a flowchart of the experimental procedure used in this study. Table 1 shows the five different CuSn10 composite materials prepared here with different volume fractions of h-BN. The CuSn10 composites with 4, 7, 10, and 14 vol.% h-BN are designed as samples 1Ae4A, while sample 1B is the CuSn10 composite with 4 vol.% modified h-BN. Firstly, for samples 1Ae4A, mixtures of CuSn10, h-BN, and the strengthening elements were prepared by planetary ball milling at about 80 rpm for 48 h with a mass ratio of balls to powder of 6:1. Then the mixtures were bidirectionally pressed at room temperature under 500 MPa. The obtained pellets were sintered under a hydrogen atmosphere at 600 C for 60 min and then at 1100 C for 90 min with a heating rate of 10 C/min. For sample 1B, the CuSn10, 4 vol%h-BN, strengthening elements, and 1.5 wt.% APAM and 15 wt.% water were mixed for 48 h in a planetary ball mill with a mass ratio of balls to powder of 6:1. After drying at 80 C for 8 h, the obtained powders underwent the same compaction and sintering process as samples 1Ae4A.
Table 1 Chemical compositions of CuSn10/h-BN composites (vol.%). Sample designation
h-BN
Ni þ Fe þ Mn
WC
CuSn10
1A 2A 3A 4A 1B
4 7 10 14 4
18 18 18 18 18
1 1 1 1 1
Balance Balance Balance Balance Balance
The morphology and composition of the samples were characterized using field-emission scanning electron microscopy (SEM) using an instrument equipped with an energy-dispersive X-ray spectrometer (EDS). The X-ray diffraction (XRD) analysis was used to identify the crystal structure of the pristine and modified h-BN powders. Electron probe microanalysis (EPMA) was used to investigate the dispersion of h-BN. Fourier transform infrared (FT-IR) spectrometry was performed using a Thermo Scientific Nicolet 6700 instrument and the KBr pellet method in the range of 500e4400 cm1 with a resolution of 4 cm1. The chemical composition and functional groups were characterized by X-ray photoelectron spectroscopy (XPS; Thermo Scientific K-Alpha) at 300 W (Al K radiation). The zeta potential of untreated and modified h-BN powders in an alcoholic solution was measured with a Zetazier 3000HSA instrument. Wet grinding of h-BN modified by APAM was carried out by high-energy ball milling. Pristine h-BN with APAM in concentrations of 0, 1, 1.5, 2, and 2.5 wt.% and 15 wt.% deionized water were mixed for 24 h. After drying the wet mixture, the samples were homogeneously dispersed in alcohol by sonication for 1 h. Vickers microhardness measurements of the obtained specimens were determined on the polished surface considering an average of five indentations for each specimen using Vickers indentation with an indentation load of 4.9 N for 10 s. Tensile tests were carried out according to ASTM D638 with a crosshead speed of 2 mm/min and a gauge length of 50 mm. Flexural testing of the composites was conducted using a universal testing machine in three-point bending mode at a constant speed of 2.0 mm/min and a span length of 32 mm. 3. Results and discussion 3.1. Powder characterization
2.2. Characterization and testing The relative bulk density was calculated from the ratio of the bulk density to the theoretical density. The porosity was calculated from the density differences measured using Archimedes principle.
SEM micrographs of CuSn10 and h-BN powders are presented in Fig. 2. The morphology of the CuSn10 powders showed fine spherical particles, while the h-BN powders had a fine flake-like structure with a particle size ranging between 10 and 15 mm.
Fig. 1. Flowchart of the experimental procedure used in this study.
T. Li et al. / Journal of Alloys and Compounds 723 (2017) 345e353
347
Fig. 2. SEM images of the investigated powders. (a) CuSn10 and (b) h-BN.
Fig. 3. SEM images of CuSn10/h-BN composites.(a) 4 vol.% h-BN, (b) 7 vol.% h-BN, (c) 10 vol.% h-BN, and (d) 14 vol.% h-BN. (e) SEM image of EDS line scan and (f) the corresponding elemental concentration as a function of distance across the pore outlined in (b).
348
T. Li et al. / Journal of Alloys and Compounds 723 (2017) 345e353
Table 2 Physical and mechanical properties of the sintered samples 1Ae4A. Sample Relative density Porosity Hardness Tensile Strength Bending Strength (%) (%) (HV) (MPa) (MPa) 1A 2A 3A 4A
80.5 78 76.1 71
20.1 20.4 23.5 27.7
157 146 91 62
134 114 106 33
238 176 111 67
3.2. Microstructure, physical and mechanical properties of sintered samples 1Ae4A The microstructures of the sintered samples 1Ae4A are shown in Fig. 3(aed). The black areas in the photograph are thought to be pores. These figures show that the least pores (highest matrix density) appeared in Fig. 3a. With increasing h-BN content, the density of the samples gradually decreased. Fig. 3e shows the position of an EDS line scan across a pore (this analysis region is indicated by the square in Fig. 3b) where Fig. 3f shows that the corresponding elemental concentrations; It can be seen that the content of B and N were highest within the pore, while the content of Cu was the lowest in the pore (that is, h-BN was mainly present in the pores) [11]. The physical and mechanical properties of the sintered samples 1Ae4A are shown in Table 2. The relative densities of the CuSn10 composites containing 4, 7, 10, and 14 vol.% h-BN were 80.5%, 78%, 76.1%, and 71%, respectively. This agrees with the hypothesis that the density of the specimens will decrease with increasing h-BN content (and corresponding decrease in Cu content), as the density of h-BN (2.1 g/cm3) is lower than that of Cu (8.9 g/cm3) [8]. It is evident that the mechanical properties of CuSn10/h-BN are directly proportional to the h-BN content. In general, the incorporation of softer h-BN powders in the
CuSn10 matrix resulted in a decrease in the hardness. Likewise, the tensile strength and bending strength of the sintered specimens was inversely proportional to the h-BN content. The poor affinity between the h-BN and matrix gives rise to weak interfaces which consequently degrade the mechanical properties. It was clear that sample 1A showed the best overall performance. During sintering, some of the h-BN powder reacted with the matrix powder, forming the boride phase at the particle boundaries. Unreacted or remaining h-BN powder, in the form of softphase particles embedded in the composite, acted as a solid lubricant. This phenomenon is beneficial for the lubricating function of the composites [9]. When a low concentration of pure h-BN was added, the compacted copper matrix could appropriately soften and melt with heating, resulting in the sintering neck growing and spreading. This allowed the copper matrix particles to merge and sinter, resulting in shrinkage and elimination of some of the pores. However, larger concentrations of h-BN resulted in agglomeration, hindering the sintering process and the formation an incomplete interface structure; the interface between copper particles contained many pores, resulting in a weakened interface structure and poor physical and mechanical properties. With increasing h-BN content, this negative effect became more noticeable. Therefore, the highest densification was observed at 4 vol.% h-BN, with relatively strong bonding between copper particles and fewer pores. A schematic of the sintering process is shown in Fig. 4. 3.3. Surface modification of h-BN The presence of the polymer and the surfactant affects the charge of slipping plane, which is expressed in the specific value of electrokinetic zeta potential. Fig. 5 shows the zeta potential of h-BN with different mass fractions of APAM. The presence of APAM resulted in a significant improvement of the system dispersibility. The adsorption of APAM on the surface of h-BN resulted in a
Fig. 4. Schematic of the sintering process of the CuSn10/h-BN composite.
T. Li et al. / Journal of Alloys and Compounds 723 (2017) 345e353
Fig. 5. Zeta potential of h-BN with different mass fractions of APAM.
Fig. 6. XRD patterns of h-BN and h-BN-APAM.
significant reduction of the zeta potential compared to the untreated h-BN (37.5 mV), with the lowest value of 53.5 mV observed when the mass fraction of APAM was 1.5%. However, the zeta potential gradually increased when the concentration of APAM was more than 1.5 wt.%; the zeta potential of h-BN with 2.5 wt.% APAM was 17.7 mV, which was much higher than that of pristine h-BN. Fig. 6 compares the XRD patterns of pristine h-BN and APAM surface-modified h-BN. No significant changes in the peak position and intensity or crystal phase were observed before and after surface modification. This means that the crystal structure of the h-BN particles was not affected by the surface modification treatment. It is known that h-BN particles have a plate-like shape where the flat surfaces correspond to the basal planes of the hexagonal crystal structure. The basal plane, crystallographically denoted as (0001), is molecularly smooth and has no surface functional groups available for chemical bonding or interaction. However, the edge planes of the platelets do have functional groups, such as hydroxyl group and amino groups [25]. FT-IR spectra of pristine h-BN and h-BN modified with APAM are shown in Fig. 7. In both these curves, the strong absorption
349
Fig. 7. FT-IR spectra of h-BN and h-BN-APAM.
peaks at approximately 1373 and 818 cm1 were assigned to BeN stretching and BeN bending, respectively; the broad band around 3417 cm1 of pristine h-BN was attributed to the surface eNH2 stretching vibrations [25,26]. These functional groups are mainly bound covalently to boron atoms in the stacking planes of a hexagonal network structure with alternately placed boron and nitrogen atoms [25]. After the modification with APAM, the eNH2 band disappeared, which may imply that the amide groups on the h-BN surface were chemically reacted with carboxyl groups in the APAM. Moreover, the additional peak located at 636 cm1 in the APAM-modified h-BN spectrum was ascribed to the of SO2 4 vibration. SEM and AFM images were used to investigate the differences in surface morphology between pristine and modified h-BN. As shown in Fig. 8b, there were many protrusions distributed on the surface of modified h-BN, while the surface of the pristine h-BN (Fig. 8a and c) was smooth. The AFM image of the modified h-BN shown in Fig. 8d is also a good illustration of this. In other words, the particles after modification become rough with a granular surface layer, attributed to the APAM on the surface of the modified h-BN particles. EDS analyses showed the existence of B, N, C, O, Na, and Pt, where the Na was from the APAM [21], proving the existence of APAM on the h-BN surface. The Pt was deposited on the sample surface to enhance the conductivity for SEM experiments. The chemical changes by the surface modification of the h-BN will be discussed considering the following XPS data. The XPS technique, which is highly sensitive to the surface chemical composition and chemical environment of elements in a material, was used to survey the typical chemistry and chemical states of the samples. Fig. 9a shows the Na 1s XPS curves of the samples. The Na 1s peak was located around 1072 eV in the h-BN-APAM sample, while this peak was not observed for pristine h-BN; this agreed well with the EDS data shown in Fig. 8b. Fig. 9b and c shows high resolution XPS spectra of the B 1s and C 1s regions, respectively. The B 1s peak for the pristine h-BN sample was located at 190.61 eV, while this peak was slightly shifted to a lower binding energy of 190.39 eV for the modified h-BN; this shift was attributed to the chemical changes occurring on the h-BN surface from surface treatment [27]. The carbon peak around 284.5 eV was observed for both the modified and pristine h-BN, which could be attributed to the CeC bonding. This may be due to residual carbon from the sample and
350
T. Li et al. / Journal of Alloys and Compounds 723 (2017) 345e353
Fig. 8. (a, b) SEM and (c, d) AFM images of (a, c) h-BN and (b, d) h-BN-APAM.
Fig. 9. High-resolution XPS of (a) Na 1s (b) B 1s (c) C 1s peaks of h-BN and h-BN-APAM.
T. Li et al. / Journal of Alloys and Compounds 723 (2017) 345e353
Fig. 10. Microstructure and EPMA elemental mapping images of sintered composites. Sample (a) (c) (e) 1A and (b) (d) (f) 1B.
351
352
T. Li et al. / Journal of Alloys and Compounds 723 (2017) 345e353
Fig. 11. Mechanical properties of sample 1A and 1B.
adventitious hydrocarbons from the XPS instrument itself. The intensity of the main C 1s peak was much higher in h-BN-APAM than the pristine h-BN due to the long main carbon chain length of APAM [19]. After modification with APAM, deconvolution of the C 1s spectra of h-BN-APAM yielded three Gaussian peaks, at 284.5, 285.4 and 288 eV, which could be related to CeC, CeN, and C]O bonds, respectively [19,28]; C]O is only associated with APAM. This indicates that the long APAM chains coated the h-BN surface. 3.4. Microstructure, physical and mechanical properties of sintered samples 1A and 1B Fig. 10a and b shows SEM images comparing the microstructure of sintered samples 1A and 1B. Sample 1B showed good densification, apart from a small number of homogeneously distributed pores, indicating good interfacial adhesion between h-BN and the matrix. Its porosity was 5.5%, a reduction of about 72.6% compared to that of sample 1A. The locations of the copper matrix and h-BN in these two samples were confirmed by electron probe microanalysis (EPMA) (Fig. 10c, d, e and f). It was clear that the h-BN particles showed less agglomeration in the matrix (Fig. 10f) than the pristine h-BN particles (Fig. 10e), meaning that the modification process was successfully employed to achieve a dense composite. However, the PM method cannot totally eliminate the presence of unreacted h-BN powder at the grain boundaries and in thin boride layers [9]. Comparison of the mechanical properties of samples 1A and 1B is shown in Fig. 11. It is worth noting that the hardness increased by 4.1% (from 146 to 152 HV), the tensile strength increased by 9.7%
(from 134 to 147 MPa), and the bending strength increased significantly by 47.5% (from 238 to 351 MPa) after APAM treatment. External forces can generate stress concentrations at the pores of the samples which are usually a source of cracks and can promote crack propagation [11]. When cracks encounter the h-BN particles in the propagation process, they are prone to expand rapidly because of the low binding strength between the h-BN and copper alloy matrix. Therefore, decreasing the porosity of sample 1B would enhance its mechanical properties. Fig. 12 shows the SEM fracture morphology of samples 1A and 1B. The bright part in the middle of sample 1A in Fig. 12a is a mass of h-BN, which inhibited the grain growth of copper particles and the excessive amounts agglomerated in the form of pockets including pores. This poor bonding between the h-BN and matrix led to various pores, which were considered weak regions. However, as show in Fig. 12b, the flaky h-BN was homogeneously inserted at the edge of the matrix, where the regions 2, 3, and 4 indicated on the image were selected for EDS analysis. Table 3 shows the mass percentage of elements calculated in these four regions; the contents of B and N were mainly at the boundaries of the copper matrix, reflecting good cohesion between the h-BN and matrix. During the surface treatment with APAM, two cross-linking mechanisms occurred, physical entanglement and chemical crosslinking [29]. On one hand, the most distinguishing feature of the APAM gel is the inclusion of all the components, where all the powders were allowed to evenly disperse within the threedimensionally cross-linked network system. On the other hand, the active edge planes of the h-BN most probably interacted chemically with the surrounding long APAM chains; the reactions between the amino groups on the end of h-BN and carboxyl in the APAM chains are useful in homogeneously dispersing h-BN in the matrix. In other words, APAM can act as a bridge between the matrix and h-BN to enhance their affinity, thereby improving the performance of the composite. In addition, the APAM can be decomposed completely during the sintering process without leaving impurities [30]. This possible mechanism is shown in Fig. 13.
Table 3 Elemental composition from EDS analysis of the areas defined in Fig. 12 (mass %). Area
B
N
C
O
Fe
Ni
Cu
Sn
W
1 2 3 4
47.36 32.25 31.45 2.41
29.34 8.23 13.67 0.48
11.73 19.02
3.68 15.66 6.14
1.37
1.51 1.68 7.05 8.48
1.85 18.76 37.68 73.15
3.11 3.67 2.35 9.84
0.05 0.73 1.66
Fig. 12. SEM fractographs of sample (a) 1A and (b) 1B.
5.55
0.09
T. Li et al. / Journal of Alloys and Compounds 723 (2017) 345e353
353
Fig. 13. Schematic of the function of APAM in the CuSn10/h-BN composite.
4. Conclusions Here, the highest relative density, hardness, tensile strength, and bending strength of the sintered body were obtained for a composite with 4 vol.% pristine h-BN; the porosity was 20.1%, the hardness was 157 HV, and the tensile strength and bending strength were 147 MPa and 238 MPa, respectively. Long APAM chains could be introduced on the h-BN surface without altering its crystal configuration. The reactions between the amino groups on the end of h-BN and carboxyl in the APAM chains facilitated the homogeneous dispersion of h-BN in the matrix. The modified h-BN exhibited less agglomeration in the matrix than pristine h-BN due to the effect of physical entanglement and chemical bonding between h-BN and APAM. Compared with the composite CuSn10-4 vol%h-BN, the porosity of the same composite with APAM-modified h-BN reduced from 20.1% to 5.5%, the tensile strength increased from 134 to 147 MPa, and the bending strength significantly increased from 238 to 351 MPa, respectively. From these results, it was concluded that using APAM to modify the h-BN is a promising method for preparing dense tin bronze/h-BN composites using a PM route. References [1] S. Zhang, J. Zhou, B. Guo, et al., Friction and wear behavior of laser cladding Ni/ h-BN self-lubricating composite coating, Mater. Sci. Eng. A 491 (1e2) (2008) 47e54. [2] H.M. Mallikarjuna, K.T. Kashyap, P.G. Koppad, et al., Microstructure and dry sliding wear behavior of Cu-Sn alloy reinforced with multiwalled carbon nanotubes, Trans. Nonferrous Metals Soc. China 26 (7) (2016) 1755e1764. [3] H. Kato, M. Takama, Y. Iwai, et al., Wear and mechanical properties of sintered copperetin composites containing graphite or molybdenum disulfide, Wear 255 (1e6) (2003) 573e578. [4] S. Chen, et al., Preparation of novel polytetrafluoroethylene/copper-matrix self-lubricating composite materials, J. Compos. Mater. 48 (13) (2014) 1561e1574. [5] J.L. Li, D.S. Xiong, Tribological properties of nickel-based self-lubricating composite at elevated temperature and counterface material selection, Wear 265 (3e4) (2008) 533e539. [6] R. Tyagi, D. Xiong, J. Li, Effect of load and sliding speed on friction and wear behavior of silver/h-BN containing Ni-base P/M composites, Wear 270 (7e8) (2011) 423e430. [7] X.F. Wei, R.C. Wang, Y. Feng, et al., Effects of h-BN content on properties of NiCr/h-BN composite, J. Central South Univ. 18 (5) (2011) 1334e1339. [8] O.A.M. Elkady, A. Abu-Oqail, E.M.M. Ewais, et al., Physico-mechanical and tribological properties of Cu/h-BN nanocomposites synthesized by PM route, J. Alloys Compd. 625 (2015) 309e317. [9] S. Mahathanabodee, T. Palathai, S. Raadnui, et al., Effects of hexagonal boron nitride and sintering temperature on mechanical and tribological properties of SS316L/h-BN composites, Mater. Des. 46 (4) (2013) 588e597. [10] Y. Kimura, T. Wakabayashi, K. Okada, et al., Boron nitride as a lubricant
additive, Wear 232 (2) (1999) 199e206. [11] S. Chen, Y. Bi, H. Zhang, et al., Effect of BN fraction on the mechanical and tribological properties of Cu alloy/BN self-lubricating sleeves, J. Compos. Mater. 49 (30) (2015). [12] R. Haubner, M. Wilhelm, R. Weissenbacher, et al., Boron Nitrides d Properties, Synthesis and Applications[M]//High Performance Non-oxide Ceramics II, Springer Berlin Heidelberg, 2002, pp. 1e45. [13] B. Chen, Q. Bi, J. Yang, et al., Tribological properties of solid lubricants (graphite, h-BN) for Cu-based P/M friction composites, Tribol. Int. 41 (12) (2008) 1145e1152. [14] D. Fan, J. Feng, S. Zhang, et al., Synthesis, structure, and magnetic properties of Ni and Co nanoparticles encapsulated by few-layer h-BN, J. Alloys Compd. 689 (2016) 153e160. [15] K.S. Lee, Y.S. Kim, M. Tosa, et al., Mechanical properties of hexagonal boron nitride synthesized from film of Cu/BN mixture by surface segregation, Appl. Surf. Sci. 169 (00) (2001) 420e424. [16] G. Hammes, R. Schroeder, C. Binder, et al., Effect of double pressing/double sintering on the sliding wear of self-lubricating sintered composites, Tribol. Int. 2014 (70) (February 2014) 119e127. [17] S. Lan, L. Li, D. Xu, et al., Surface modification of magnesium hydroxide using vinyltriethoxysilane by dry process, Appl. Surf. Sci. 382 (2016) 56e62. [18] X. Shang, Y. Zhu, Z. Li, Surface modification of silicon carbide with silane coupling agent and hexadecyl iodiele, Appl. Surf. Sci. 394 (2016) 169e177. [19] I. Jang, K.H. Shin, I. Yang, et al., Enhancement of Thermal Conductivity of BN/Epoxy Composite through Surface Modification with Silane Coupling Agents, 2017. [20] X.W. Li, R.G. Song, Y. Jiang, et al., Surface modification of TiO2, nanoparticles and its effect on the properties of fluoropolymer/TiO2, nanocomposite coatings, Appl. Surf. Sci. 276 (3) (2013) 761e768. [21] Małgorzata Wisniewska, Stanisław Chibowski, Teresa Urban, Modification of the alumina surface properties by adsorbed anionic polyacrylamidedimpact of polymer hydrolysis, J. Industrial Eng. Chem. 21 (1) (2014) 925e931. [22] X.C. Zhang, W.P. Miller, Polyacrylamide effect on infiltration and erosion in furrows, Soil Sci. Soc. Am. J. 60 (3) (1996) 866e872. [23] E. Pefferkorn, Polyacrylamide at solid/liquid interfaces, J. Colloid & Interface Sci. 216 (2) (1999) 197e220. [24] E. Pefferkorn, Structure and stability of natural organic matter/soil complexes and related synthetic and mixed analogues, Adv. Colloid & Interface Sci. 73 (1997) 127e200. [25] K. Sato, H. Horibe, T. Shirai, et al., Thermally conductive composite films of hexagonal boron nitride and polyimide with affinity-enhanced interfaces, J. Mater. Chem. 20 (14) (2010) 2749e2752. [26] Y. Gao, A. Gu, Y. Jiao, et al., High-performance hexagonal boron nitride/bismaleimide composites with high thermal conductivity, low coefficient of thermal expansion, and low dielectric loss, Polym. Adv. Technol. 23 (5) (2012) 919e928. [27] P. Kidambi, R. Blume, J. Kling, et al., Situ observations during chemical vapor depositionof hexagonal boron nitride on polycrystalline copper, Chem. Mater. A Publ. Am. Chem. Soc. 26 (22) (2014) 6380. [28] Y. Hua, C. Wang, Z. Shan, et al., Effect of surface modification on carbon fiber and its reinforced phenolic matrix composite, Appl. Surf. Sci. 259 (41) (2012) 288e293. [29] Z.Q. Duan, M. Zhong, F.K. Shi, et al., Transparent h-BN/polyacrylamide nanocomposite hydrogels with enhanced mechanical properties, Chin. Chem. Lett. 27 (9) (2016) 1490e1494. [30] M. Tahmasebpour, A.A. Babaluo, M.K. Aghjeh, Synthesis of zirconia nanopowders from various zirconium salts via polyacrylamide gel method, J. Eur. Ceram. Soc. 28 (4) (2008) 773e778.
Contents lists available at ScienceDirect
Journal of Alloys and Compounds journal homepage: http://www.elsevier.com/locate/jalcom
Surface modification of h-BN and its influence on the mechanical properties of CuSn10/h-BN composites Ting Li a, b, Danqing Yi a, b, Jun Hu a, b, Jiao Xu a, b, Junlei Liu a, b, Bin Wang a, b, * a b
School of Material Science and Engineering, Central South University, Changsha, Hunan 410083, PR China Light Alloy Research Institute, Central South University, Changsha, Hunan 410083, PR China
a r t i c l e i n f o
a b s t r a c t
Article history: Received 14 April 2017 Received in revised form 23 June 2017 Accepted 24 June 2017 Available online 27 June 2017
CuSn10 matrix composites with hexagonal boron nitride (h-BN) in concentrations of 4, 7, 10, and 14 vol.% were prepared using a powder metallurgy process. The effect of modifying the h-BN with anionic polyacrylamide (APAM) by high-energy ball milling was investigated. The bonding mechanism between h-BN and APAM was characterized by Fourier transform infrared spectroscopy (FT-IR), X-ray photoelectron spectroscopy (XPS), and X-ray diffraction (XRD). The microstructure, hardness, tensile strength, and bending strength of sintered composites were investigated. The composite CuSn10-4 vol%h-BN showed the lowest porosity of 20.1%. The long APAM chains could be introduced on the h-BN surface without altering its crystal structure. The surface modified h-BN exhibited less agglomeration in the matrix than pristine h-BN, due to the physical entanglement and chemical bonding between h-BN and APAM. Comparing the composites with 4 vol.% h-BN, without and with APAM modification, a reduction in the porosity from 20.1% to 5.5%, and increase in the tensile strength from 134 to 147 MPa, and a significantly increased in the bending strength from 238 to 351 MPa, respectively, were observed. © 2017 Published by Elsevier B.V.
Keywords: h-BN Anionic polyacrylamide Surface modification Copper matrix composite Powder metallurgy
1. Introduction Many tribological system working in high-temperature aggressive environments demand combinations of a low friction coefficient, good high-temperature oxidation resistance, and good wear resistance [1]. Tin bronze matrix powder metallurgy (PM) composites with excellent tribological properties and good heat conductivity and corrosion resistance have been widely investigated [2e4]. Traditional solid lubricants, such as graphite and MoS2, are good choices for tin bronze friction materials, where their lamellar structures are beneficial for decreasing friction, but most such lubricants can only be used under ambient temperature [3,5e7]. Hexagonal boron nitride (h-BN), is an inorganic analogue of graphene and has a lamellar structure consisting of a stack of hexagonal sheets with a strong covalent bond between boron and nitrogen atoms. The sheets are held together by weak van der Waals forces that allow shearing when force is applied parallel to the sheets. Thus, it provides the expected friction reduction and
* Corresponding author. School of Material Science and Engineering, Central South University, Changsha, Hunan 410083, PR China. E-mail address: [email protected] (B. Wang). http://dx.doi.org/10.1016/j.jallcom.2017.06.264 0925-8388/© 2017 Published by Elsevier B.V.
results in very efficient lubrication over a wide temperature range [1,8e13]. However, due to the surface incompatibility between hBN and the matrix alloy, the physical mixing of h-BN particles leads to phase segregation between the h-BN and matrix, giving rise to agglomeration of h-BN particles, which may negatively affect the physical and mechanical properties of the resulting composites [1,14e16]. Thus, reducing the agglomeration of h-BN particles and improving the interfacial adhesion between h-BN and the matrix surface have been the most important issues for the development of a tin bronze/h-BN composite. To break down the particle agglomerates and improve the dispersibility of the particles in the matrix, surface modification is an effective and commonly used method [17]. Silicon carbide has previously been modified by a silane coupling agent and hexadecyl iodiele [18]. Silane coupling agents with different carbon chains (C3 and C16) were introduced on the BN surface to improve the affinity of BN for the epoxy resin [19]. In the work of Li et al., TiO2 was modified by KH550 through a silanization reaction [20]. Anionic polyacrylamide (APAM) is produced during the polymerization reaction of acrylamide and has both amide and carboxyl functional groups [21]. APAM gel is a typical soft material that has a unique, three-dimensionally cross-linked network structure swollen with a large amount of water. The unique structure of these
346
T. Li et al. / Journal of Alloys and Compounds 723 (2017) 345e353
systems means they can be safely applied in materials processing and hence the study of these materials is of high utility. Currently, APAM is mainly applied in cosmetics, pharmaceutics and paint production, minerals flotation, oil recovery, agriculture, and environmental protection applications [22e24]. In order to realize homogeneous dispersion of h-BN in a tin bronze matrix, here we propose to prepare CuSn10/h-BN composites using APAM as a surface modifier. The modification mechanism of APAM on the surface of h-BN was studied. In addition, the influence of the APAM on the mechanical properties of the composite made with the modified h-BN was investigated. 2. Experimental 2.1. Materials and processing CuSn10 powder with an average particle size ~35 mm with a purity of 99.99% was used as the metal matrix material. Commercial h-BN (99.9% purity) was used as a lubrication material while the other powders (Ni, Fe, Mn, and WC) were used as reinforcement materials. All these powders were passed through a 200 mesh sieve. APAM has the chemical formula eCH2eCH(CONH2) eCH2eCH(COONa)e and an average molecular weight of 14000000. Fig. 1 shows a flowchart of the experimental procedure used in this study. Table 1 shows the five different CuSn10 composite materials prepared here with different volume fractions of h-BN. The CuSn10 composites with 4, 7, 10, and 14 vol.% h-BN are designed as samples 1Ae4A, while sample 1B is the CuSn10 composite with 4 vol.% modified h-BN. Firstly, for samples 1Ae4A, mixtures of CuSn10, h-BN, and the strengthening elements were prepared by planetary ball milling at about 80 rpm for 48 h with a mass ratio of balls to powder of 6:1. Then the mixtures were bidirectionally pressed at room temperature under 500 MPa. The obtained pellets were sintered under a hydrogen atmosphere at 600 C for 60 min and then at 1100 C for 90 min with a heating rate of 10 C/min. For sample 1B, the CuSn10, 4 vol%h-BN, strengthening elements, and 1.5 wt.% APAM and 15 wt.% water were mixed for 48 h in a planetary ball mill with a mass ratio of balls to powder of 6:1. After drying at 80 C for 8 h, the obtained powders underwent the same compaction and sintering process as samples 1Ae4A.
Table 1 Chemical compositions of CuSn10/h-BN composites (vol.%). Sample designation
h-BN
Ni þ Fe þ Mn
WC
CuSn10
1A 2A 3A 4A 1B
4 7 10 14 4
18 18 18 18 18
1 1 1 1 1
Balance Balance Balance Balance Balance
The morphology and composition of the samples were characterized using field-emission scanning electron microscopy (SEM) using an instrument equipped with an energy-dispersive X-ray spectrometer (EDS). The X-ray diffraction (XRD) analysis was used to identify the crystal structure of the pristine and modified h-BN powders. Electron probe microanalysis (EPMA) was used to investigate the dispersion of h-BN. Fourier transform infrared (FT-IR) spectrometry was performed using a Thermo Scientific Nicolet 6700 instrument and the KBr pellet method in the range of 500e4400 cm1 with a resolution of 4 cm1. The chemical composition and functional groups were characterized by X-ray photoelectron spectroscopy (XPS; Thermo Scientific K-Alpha) at 300 W (Al K radiation). The zeta potential of untreated and modified h-BN powders in an alcoholic solution was measured with a Zetazier 3000HSA instrument. Wet grinding of h-BN modified by APAM was carried out by high-energy ball milling. Pristine h-BN with APAM in concentrations of 0, 1, 1.5, 2, and 2.5 wt.% and 15 wt.% deionized water were mixed for 24 h. After drying the wet mixture, the samples were homogeneously dispersed in alcohol by sonication for 1 h. Vickers microhardness measurements of the obtained specimens were determined on the polished surface considering an average of five indentations for each specimen using Vickers indentation with an indentation load of 4.9 N for 10 s. Tensile tests were carried out according to ASTM D638 with a crosshead speed of 2 mm/min and a gauge length of 50 mm. Flexural testing of the composites was conducted using a universal testing machine in three-point bending mode at a constant speed of 2.0 mm/min and a span length of 32 mm. 3. Results and discussion 3.1. Powder characterization
2.2. Characterization and testing The relative bulk density was calculated from the ratio of the bulk density to the theoretical density. The porosity was calculated from the density differences measured using Archimedes principle.
SEM micrographs of CuSn10 and h-BN powders are presented in Fig. 2. The morphology of the CuSn10 powders showed fine spherical particles, while the h-BN powders had a fine flake-like structure with a particle size ranging between 10 and 15 mm.
Fig. 1. Flowchart of the experimental procedure used in this study.
T. Li et al. / Journal of Alloys and Compounds 723 (2017) 345e353
347
Fig. 2. SEM images of the investigated powders. (a) CuSn10 and (b) h-BN.
Fig. 3. SEM images of CuSn10/h-BN composites.(a) 4 vol.% h-BN, (b) 7 vol.% h-BN, (c) 10 vol.% h-BN, and (d) 14 vol.% h-BN. (e) SEM image of EDS line scan and (f) the corresponding elemental concentration as a function of distance across the pore outlined in (b).
348
T. Li et al. / Journal of Alloys and Compounds 723 (2017) 345e353
Table 2 Physical and mechanical properties of the sintered samples 1Ae4A. Sample Relative density Porosity Hardness Tensile Strength Bending Strength (%) (%) (HV) (MPa) (MPa) 1A 2A 3A 4A
80.5 78 76.1 71
20.1 20.4 23.5 27.7
157 146 91 62
134 114 106 33
238 176 111 67
3.2. Microstructure, physical and mechanical properties of sintered samples 1Ae4A The microstructures of the sintered samples 1Ae4A are shown in Fig. 3(aed). The black areas in the photograph are thought to be pores. These figures show that the least pores (highest matrix density) appeared in Fig. 3a. With increasing h-BN content, the density of the samples gradually decreased. Fig. 3e shows the position of an EDS line scan across a pore (this analysis region is indicated by the square in Fig. 3b) where Fig. 3f shows that the corresponding elemental concentrations; It can be seen that the content of B and N were highest within the pore, while the content of Cu was the lowest in the pore (that is, h-BN was mainly present in the pores) [11]. The physical and mechanical properties of the sintered samples 1Ae4A are shown in Table 2. The relative densities of the CuSn10 composites containing 4, 7, 10, and 14 vol.% h-BN were 80.5%, 78%, 76.1%, and 71%, respectively. This agrees with the hypothesis that the density of the specimens will decrease with increasing h-BN content (and corresponding decrease in Cu content), as the density of h-BN (2.1 g/cm3) is lower than that of Cu (8.9 g/cm3) [8]. It is evident that the mechanical properties of CuSn10/h-BN are directly proportional to the h-BN content. In general, the incorporation of softer h-BN powders in the
CuSn10 matrix resulted in a decrease in the hardness. Likewise, the tensile strength and bending strength of the sintered specimens was inversely proportional to the h-BN content. The poor affinity between the h-BN and matrix gives rise to weak interfaces which consequently degrade the mechanical properties. It was clear that sample 1A showed the best overall performance. During sintering, some of the h-BN powder reacted with the matrix powder, forming the boride phase at the particle boundaries. Unreacted or remaining h-BN powder, in the form of softphase particles embedded in the composite, acted as a solid lubricant. This phenomenon is beneficial for the lubricating function of the composites [9]. When a low concentration of pure h-BN was added, the compacted copper matrix could appropriately soften and melt with heating, resulting in the sintering neck growing and spreading. This allowed the copper matrix particles to merge and sinter, resulting in shrinkage and elimination of some of the pores. However, larger concentrations of h-BN resulted in agglomeration, hindering the sintering process and the formation an incomplete interface structure; the interface between copper particles contained many pores, resulting in a weakened interface structure and poor physical and mechanical properties. With increasing h-BN content, this negative effect became more noticeable. Therefore, the highest densification was observed at 4 vol.% h-BN, with relatively strong bonding between copper particles and fewer pores. A schematic of the sintering process is shown in Fig. 4. 3.3. Surface modification of h-BN The presence of the polymer and the surfactant affects the charge of slipping plane, which is expressed in the specific value of electrokinetic zeta potential. Fig. 5 shows the zeta potential of h-BN with different mass fractions of APAM. The presence of APAM resulted in a significant improvement of the system dispersibility. The adsorption of APAM on the surface of h-BN resulted in a
Fig. 4. Schematic of the sintering process of the CuSn10/h-BN composite.
T. Li et al. / Journal of Alloys and Compounds 723 (2017) 345e353
Fig. 5. Zeta potential of h-BN with different mass fractions of APAM.
Fig. 6. XRD patterns of h-BN and h-BN-APAM.
significant reduction of the zeta potential compared to the untreated h-BN (37.5 mV), with the lowest value of 53.5 mV observed when the mass fraction of APAM was 1.5%. However, the zeta potential gradually increased when the concentration of APAM was more than 1.5 wt.%; the zeta potential of h-BN with 2.5 wt.% APAM was 17.7 mV, which was much higher than that of pristine h-BN. Fig. 6 compares the XRD patterns of pristine h-BN and APAM surface-modified h-BN. No significant changes in the peak position and intensity or crystal phase were observed before and after surface modification. This means that the crystal structure of the h-BN particles was not affected by the surface modification treatment. It is known that h-BN particles have a plate-like shape where the flat surfaces correspond to the basal planes of the hexagonal crystal structure. The basal plane, crystallographically denoted as (0001), is molecularly smooth and has no surface functional groups available for chemical bonding or interaction. However, the edge planes of the platelets do have functional groups, such as hydroxyl group and amino groups [25]. FT-IR spectra of pristine h-BN and h-BN modified with APAM are shown in Fig. 7. In both these curves, the strong absorption
349
Fig. 7. FT-IR spectra of h-BN and h-BN-APAM.
peaks at approximately 1373 and 818 cm1 were assigned to BeN stretching and BeN bending, respectively; the broad band around 3417 cm1 of pristine h-BN was attributed to the surface eNH2 stretching vibrations [25,26]. These functional groups are mainly bound covalently to boron atoms in the stacking planes of a hexagonal network structure with alternately placed boron and nitrogen atoms [25]. After the modification with APAM, the eNH2 band disappeared, which may imply that the amide groups on the h-BN surface were chemically reacted with carboxyl groups in the APAM. Moreover, the additional peak located at 636 cm1 in the APAM-modified h-BN spectrum was ascribed to the of SO2 4 vibration. SEM and AFM images were used to investigate the differences in surface morphology between pristine and modified h-BN. As shown in Fig. 8b, there were many protrusions distributed on the surface of modified h-BN, while the surface of the pristine h-BN (Fig. 8a and c) was smooth. The AFM image of the modified h-BN shown in Fig. 8d is also a good illustration of this. In other words, the particles after modification become rough with a granular surface layer, attributed to the APAM on the surface of the modified h-BN particles. EDS analyses showed the existence of B, N, C, O, Na, and Pt, where the Na was from the APAM [21], proving the existence of APAM on the h-BN surface. The Pt was deposited on the sample surface to enhance the conductivity for SEM experiments. The chemical changes by the surface modification of the h-BN will be discussed considering the following XPS data. The XPS technique, which is highly sensitive to the surface chemical composition and chemical environment of elements in a material, was used to survey the typical chemistry and chemical states of the samples. Fig. 9a shows the Na 1s XPS curves of the samples. The Na 1s peak was located around 1072 eV in the h-BN-APAM sample, while this peak was not observed for pristine h-BN; this agreed well with the EDS data shown in Fig. 8b. Fig. 9b and c shows high resolution XPS spectra of the B 1s and C 1s regions, respectively. The B 1s peak for the pristine h-BN sample was located at 190.61 eV, while this peak was slightly shifted to a lower binding energy of 190.39 eV for the modified h-BN; this shift was attributed to the chemical changes occurring on the h-BN surface from surface treatment [27]. The carbon peak around 284.5 eV was observed for both the modified and pristine h-BN, which could be attributed to the CeC bonding. This may be due to residual carbon from the sample and
350
T. Li et al. / Journal of Alloys and Compounds 723 (2017) 345e353
Fig. 8. (a, b) SEM and (c, d) AFM images of (a, c) h-BN and (b, d) h-BN-APAM.
Fig. 9. High-resolution XPS of (a) Na 1s (b) B 1s (c) C 1s peaks of h-BN and h-BN-APAM.
T. Li et al. / Journal of Alloys and Compounds 723 (2017) 345e353
Fig. 10. Microstructure and EPMA elemental mapping images of sintered composites. Sample (a) (c) (e) 1A and (b) (d) (f) 1B.
351
352
T. Li et al. / Journal of Alloys and Compounds 723 (2017) 345e353
Fig. 11. Mechanical properties of sample 1A and 1B.
adventitious hydrocarbons from the XPS instrument itself. The intensity of the main C 1s peak was much higher in h-BN-APAM than the pristine h-BN due to the long main carbon chain length of APAM [19]. After modification with APAM, deconvolution of the C 1s spectra of h-BN-APAM yielded three Gaussian peaks, at 284.5, 285.4 and 288 eV, which could be related to CeC, CeN, and C]O bonds, respectively [19,28]; C]O is only associated with APAM. This indicates that the long APAM chains coated the h-BN surface. 3.4. Microstructure, physical and mechanical properties of sintered samples 1A and 1B Fig. 10a and b shows SEM images comparing the microstructure of sintered samples 1A and 1B. Sample 1B showed good densification, apart from a small number of homogeneously distributed pores, indicating good interfacial adhesion between h-BN and the matrix. Its porosity was 5.5%, a reduction of about 72.6% compared to that of sample 1A. The locations of the copper matrix and h-BN in these two samples were confirmed by electron probe microanalysis (EPMA) (Fig. 10c, d, e and f). It was clear that the h-BN particles showed less agglomeration in the matrix (Fig. 10f) than the pristine h-BN particles (Fig. 10e), meaning that the modification process was successfully employed to achieve a dense composite. However, the PM method cannot totally eliminate the presence of unreacted h-BN powder at the grain boundaries and in thin boride layers [9]. Comparison of the mechanical properties of samples 1A and 1B is shown in Fig. 11. It is worth noting that the hardness increased by 4.1% (from 146 to 152 HV), the tensile strength increased by 9.7%
(from 134 to 147 MPa), and the bending strength increased significantly by 47.5% (from 238 to 351 MPa) after APAM treatment. External forces can generate stress concentrations at the pores of the samples which are usually a source of cracks and can promote crack propagation [11]. When cracks encounter the h-BN particles in the propagation process, they are prone to expand rapidly because of the low binding strength between the h-BN and copper alloy matrix. Therefore, decreasing the porosity of sample 1B would enhance its mechanical properties. Fig. 12 shows the SEM fracture morphology of samples 1A and 1B. The bright part in the middle of sample 1A in Fig. 12a is a mass of h-BN, which inhibited the grain growth of copper particles and the excessive amounts agglomerated in the form of pockets including pores. This poor bonding between the h-BN and matrix led to various pores, which were considered weak regions. However, as show in Fig. 12b, the flaky h-BN was homogeneously inserted at the edge of the matrix, where the regions 2, 3, and 4 indicated on the image were selected for EDS analysis. Table 3 shows the mass percentage of elements calculated in these four regions; the contents of B and N were mainly at the boundaries of the copper matrix, reflecting good cohesion between the h-BN and matrix. During the surface treatment with APAM, two cross-linking mechanisms occurred, physical entanglement and chemical crosslinking [29]. On one hand, the most distinguishing feature of the APAM gel is the inclusion of all the components, where all the powders were allowed to evenly disperse within the threedimensionally cross-linked network system. On the other hand, the active edge planes of the h-BN most probably interacted chemically with the surrounding long APAM chains; the reactions between the amino groups on the end of h-BN and carboxyl in the APAM chains are useful in homogeneously dispersing h-BN in the matrix. In other words, APAM can act as a bridge between the matrix and h-BN to enhance their affinity, thereby improving the performance of the composite. In addition, the APAM can be decomposed completely during the sintering process without leaving impurities [30]. This possible mechanism is shown in Fig. 13.
Table 3 Elemental composition from EDS analysis of the areas defined in Fig. 12 (mass %). Area
B
N
C
O
Fe
Ni
Cu
Sn
W
1 2 3 4
47.36 32.25 31.45 2.41
29.34 8.23 13.67 0.48
11.73 19.02
3.68 15.66 6.14
1.37
1.51 1.68 7.05 8.48
1.85 18.76 37.68 73.15
3.11 3.67 2.35 9.84
0.05 0.73 1.66
Fig. 12. SEM fractographs of sample (a) 1A and (b) 1B.
5.55
0.09
T. Li et al. / Journal of Alloys and Compounds 723 (2017) 345e353
353
Fig. 13. Schematic of the function of APAM in the CuSn10/h-BN composite.
4. Conclusions Here, the highest relative density, hardness, tensile strength, and bending strength of the sintered body were obtained for a composite with 4 vol.% pristine h-BN; the porosity was 20.1%, the hardness was 157 HV, and the tensile strength and bending strength were 147 MPa and 238 MPa, respectively. Long APAM chains could be introduced on the h-BN surface without altering its crystal configuration. The reactions between the amino groups on the end of h-BN and carboxyl in the APAM chains facilitated the homogeneous dispersion of h-BN in the matrix. The modified h-BN exhibited less agglomeration in the matrix than pristine h-BN due to the effect of physical entanglement and chemical bonding between h-BN and APAM. Compared with the composite CuSn10-4 vol%h-BN, the porosity of the same composite with APAM-modified h-BN reduced from 20.1% to 5.5%, the tensile strength increased from 134 to 147 MPa, and the bending strength significantly increased from 238 to 351 MPa, respectively. From these results, it was concluded that using APAM to modify the h-BN is a promising method for preparing dense tin bronze/h-BN composites using a PM route. References [1] S. Zhang, J. Zhou, B. Guo, et al., Friction and wear behavior of laser cladding Ni/ h-BN self-lubricating composite coating, Mater. Sci. Eng. A 491 (1e2) (2008) 47e54. [2] H.M. Mallikarjuna, K.T. Kashyap, P.G. Koppad, et al., Microstructure and dry sliding wear behavior of Cu-Sn alloy reinforced with multiwalled carbon nanotubes, Trans. Nonferrous Metals Soc. China 26 (7) (2016) 1755e1764. [3] H. Kato, M. Takama, Y. Iwai, et al., Wear and mechanical properties of sintered copperetin composites containing graphite or molybdenum disulfide, Wear 255 (1e6) (2003) 573e578. [4] S. Chen, et al., Preparation of novel polytetrafluoroethylene/copper-matrix self-lubricating composite materials, J. Compos. Mater. 48 (13) (2014) 1561e1574. [5] J.L. Li, D.S. Xiong, Tribological properties of nickel-based self-lubricating composite at elevated temperature and counterface material selection, Wear 265 (3e4) (2008) 533e539. [6] R. Tyagi, D. Xiong, J. Li, Effect of load and sliding speed on friction and wear behavior of silver/h-BN containing Ni-base P/M composites, Wear 270 (7e8) (2011) 423e430. [7] X.F. Wei, R.C. Wang, Y. Feng, et al., Effects of h-BN content on properties of NiCr/h-BN composite, J. Central South Univ. 18 (5) (2011) 1334e1339. [8] O.A.M. Elkady, A. Abu-Oqail, E.M.M. Ewais, et al., Physico-mechanical and tribological properties of Cu/h-BN nanocomposites synthesized by PM route, J. Alloys Compd. 625 (2015) 309e317. [9] S. Mahathanabodee, T. Palathai, S. Raadnui, et al., Effects of hexagonal boron nitride and sintering temperature on mechanical and tribological properties of SS316L/h-BN composites, Mater. Des. 46 (4) (2013) 588e597. [10] Y. Kimura, T. Wakabayashi, K. Okada, et al., Boron nitride as a lubricant
additive, Wear 232 (2) (1999) 199e206. [11] S. Chen, Y. Bi, H. Zhang, et al., Effect of BN fraction on the mechanical and tribological properties of Cu alloy/BN self-lubricating sleeves, J. Compos. Mater. 49 (30) (2015). [12] R. Haubner, M. Wilhelm, R. Weissenbacher, et al., Boron Nitrides d Properties, Synthesis and Applications[M]//High Performance Non-oxide Ceramics II, Springer Berlin Heidelberg, 2002, pp. 1e45. [13] B. Chen, Q. Bi, J. Yang, et al., Tribological properties of solid lubricants (graphite, h-BN) for Cu-based P/M friction composites, Tribol. Int. 41 (12) (2008) 1145e1152. [14] D. Fan, J. Feng, S. Zhang, et al., Synthesis, structure, and magnetic properties of Ni and Co nanoparticles encapsulated by few-layer h-BN, J. Alloys Compd. 689 (2016) 153e160. [15] K.S. Lee, Y.S. Kim, M. Tosa, et al., Mechanical properties of hexagonal boron nitride synthesized from film of Cu/BN mixture by surface segregation, Appl. Surf. Sci. 169 (00) (2001) 420e424. [16] G. Hammes, R. Schroeder, C. Binder, et al., Effect of double pressing/double sintering on the sliding wear of self-lubricating sintered composites, Tribol. Int. 2014 (70) (February 2014) 119e127. [17] S. Lan, L. Li, D. Xu, et al., Surface modification of magnesium hydroxide using vinyltriethoxysilane by dry process, Appl. Surf. Sci. 382 (2016) 56e62. [18] X. Shang, Y. Zhu, Z. Li, Surface modification of silicon carbide with silane coupling agent and hexadecyl iodiele, Appl. Surf. Sci. 394 (2016) 169e177. [19] I. Jang, K.H. Shin, I. Yang, et al., Enhancement of Thermal Conductivity of BN/Epoxy Composite through Surface Modification with Silane Coupling Agents, 2017. [20] X.W. Li, R.G. Song, Y. Jiang, et al., Surface modification of TiO2, nanoparticles and its effect on the properties of fluoropolymer/TiO2, nanocomposite coatings, Appl. Surf. Sci. 276 (3) (2013) 761e768. [21] Małgorzata Wisniewska, Stanisław Chibowski, Teresa Urban, Modification of the alumina surface properties by adsorbed anionic polyacrylamidedimpact of polymer hydrolysis, J. Industrial Eng. Chem. 21 (1) (2014) 925e931. [22] X.C. Zhang, W.P. Miller, Polyacrylamide effect on infiltration and erosion in furrows, Soil Sci. Soc. Am. J. 60 (3) (1996) 866e872. [23] E. Pefferkorn, Polyacrylamide at solid/liquid interfaces, J. Colloid & Interface Sci. 216 (2) (1999) 197e220. [24] E. Pefferkorn, Structure and stability of natural organic matter/soil complexes and related synthetic and mixed analogues, Adv. Colloid & Interface Sci. 73 (1997) 127e200. [25] K. Sato, H. Horibe, T. Shirai, et al., Thermally conductive composite films of hexagonal boron nitride and polyimide with affinity-enhanced interfaces, J. Mater. Chem. 20 (14) (2010) 2749e2752. [26] Y. Gao, A. Gu, Y. Jiao, et al., High-performance hexagonal boron nitride/bismaleimide composites with high thermal conductivity, low coefficient of thermal expansion, and low dielectric loss, Polym. Adv. Technol. 23 (5) (2012) 919e928. [27] P. Kidambi, R. Blume, J. Kling, et al., Situ observations during chemical vapor depositionof hexagonal boron nitride on polycrystalline copper, Chem. Mater. A Publ. Am. Chem. Soc. 26 (22) (2014) 6380. [28] Y. Hua, C. Wang, Z. Shan, et al., Effect of surface modification on carbon fiber and its reinforced phenolic matrix composite, Appl. Surf. Sci. 259 (41) (2012) 288e293. [29] Z.Q. Duan, M. Zhong, F.K. Shi, et al., Transparent h-BN/polyacrylamide nanocomposite hydrogels with enhanced mechanical properties, Chin. Chem. Lett. 27 (9) (2016) 1490e1494. [30] M. Tahmasebpour, A.A. Babaluo, M.K. Aghjeh, Synthesis of zirconia nanopowders from various zirconium salts via polyacrylamide gel method, J. Eur. Ceram. Soc. 28 (4) (2008) 773e778.