Search for articles:
Min Su Park
, Jong Hak Kim
Corresponding authors:
Received: 2019-06-4
Revised: 2019-06-29
Accepted: 2019-07-8
Online: 2020-01-15
Copyright: 2020 Editorial board of Journal of Materials Science & Technology Copyright reserved, Editorial board of Journal of Materials Science & Technology
About authors:
1 These authors have contributed equally to this work.
More
Abstract
We report a composite material with the high tribological performance, which consists of one-dimensional SnO2 nanotubes (ST) and a high molecular weight poly(vinylidene fluoride) (PVDF) matrix in terms of nano-impregnation. Dissolution of PVDF in N,N-dimethylformamide (DMF) resulted in a facile penetration of PVDF into the inner hollow voids of ST, leading to the close contact. Interaction between PVDF and ST results in a beneficial effect on the chain arrangement of PVDF, providing an α-phase with better tribological property. Upon ST incorporation, the friction coefficient decreased by 85.0% to 0.408, and the specific wear rate decreased by 69.1% to 0.412, demonstrating the pivotal role of ST as a self-lubricating material due to a large interactive area and PVDF chain rearrangement.
Keywords：
From ancient times to the present, seismic issues have arisen, such as the collapse of buildings or human casualties [[1], [2], [3]]. To address these issues, tribology, which deals with the effects of friction, wear, and lubrication of interacting surfaces in relative motion, has been widely studied for several decades. To prevent critical and severe damages, several studies have attempted to develop novel materials that have both excellent frictional characteristics and robust properties. For example, carbon nanotubes/Al [4], MoS2 [5], copper matrix/graphene nanosheet [6], graphene oxide/diamond [7], and metallic glass [8] have recently been studied to reduce frictional abrasion as candidates for building ingredients. However, inorganic materials require high cost and complicated processes to fabricate composites [9]. From this point of view, polymers are used for their excellent advantages, such as low density, facile processing, and robust properties [[10],[9], [10], [11], [12]].
Among polymer candidates, polytetrafluoroethylene (PTFE) has been widely studied owing to its differentiated properties from other polymers, such as high resistance to heat and chemicals, lubricating ability, and electrical properties [[13],[1], [2], [3], [4], [5], [6], [7], [8], [9], [10], [11], [12], [13], [14], [15]]. This is a fully fluorinated PTFE, which is the reason of the above advantages; it also has disadvantages in its application in several areas, such as low processability and high price. Poly(vinylidene fluoride) (PVDF) is one of the excellent alternatives to PTFE as a partially fluorinated polymer [[16], [17], [18]]. In addition to the advantages mentioned above, it is an economical and easily processable polymer [[9],[19]].
As a supplementation of pure polymer composites, several nanostructured inorganic materials have been employed as fillers to enhance tribological performance in recent studies [20]. Because several inorganic fillers with nanostructures are beneficial for improving polymer properties in the field of tribology, various classes of inorganic fillers have been applied in polymer composites. First, carbon materials are known as anti-wear fillers, such as carbon nanotubes with one or multiple walls [[21], [22], [23]], carbon nanofibers [[24],[25]], g-C3N4 [26], etc. Second, TiO2 nanoparticles [[27],[28]], MgO nanosheet [29], and black phosphorus nanosheet [30] have also been utilized as fillers in PTFE or PVDF polymer composites. Among inorganic materials, tin (II) oxide, especially the nanotube structure, has not been used in this area to enhance frictional performance.
Here we report on the fabrication of a novel polymer/metal oxide composite utilizing a PVDF matrix and the one-dimensional tin oxide nanotubes (ST) filler to control wear behavior. Particular attention was paid to the synthesis of one-dimensional ST and the impregnation approach for PVDF/ST composites with close contact. The morphologies of ST and their composites with PVDF were characterized via differential scanning calorimetry (DSC), high-resolution X-ray diffraction (HR-XRD), and field-emission scanning electron microscopy (FE-SEM). Composite morphologies were further explored using SEM energy-dispersive X-ray spectrometry (EDS), water contact angle analysis, and Fourier transform infrared (FT-IR) spectroscopy. Scanning tunneling microscope (STM)-Micro Tribology Tester was utilized to evaluate tribological performances. We applied the recipro-mode during the wear test, which was more realistic than the circular mode for simulating the seismic imitation apparatus.
Tin (II) chloride (SnCl2, 98%), poly(vinyl pyrrolidone) (PVP, average molecular weight: ∼1,300,000 g mol-1), and poly(vinylidene fluoride) (PVDF, average Mw: ∼534,000 g mol-1) were purchased from Sigma-Aldrich. Absolute ethanol and N,N-dimethylformamide (DMF) were purchased from J.T. Baker.
First, 0.5 g of PVP and 0.4 g of SnCl2 were dissolved in a solvent mixture consisting of 2.5 ml of DMF and 2.5 ml of absolute ethanol by magnetic stirring at room temperature. After that, the homogeneous, transparent solution was poured into a needle-connected syringe. A voltage of 15 KV was applied to the needle by a direct current (DC) power, and the solution was extruded by a pump at a rate of 0.5 ml-1. PVP/SnCl2 nanofibers were electrosprayed on the collector and kept at 120 °C overnight in an oven to remove residual solvent. The completely dried powder was calcined at 500 °C for 2 h to remove the PVP and for crystallization of SnO2, resulting in the formation of SnO2 nanotube (ST).
First, different amounts of ST (0.1, 0.2, 0.3, and 0.4 g) were homogeneously dispersed in 10 ml of DMF by sonication. 0.4 g of PVDF was completely dissolved in the solution with magnetic stirring. Secondly, various amounts of PVDF (3.5, 3.4, 3.3, and 3.2 g) were uniformly dispersed in 30 ml of ethanol. The total weight of PVDF and ST for the preparation of PVDF/ST powders was fixed at 4.0 g. The PVDF/ST solution was slowly added to the PVDF-dispersed ethanol drop by drop. The solution mixture was centrifuged to separate the precipitates from the solution. The resulting precipitates were sufficiently washed with ethanol and then centrifuged several times to remove residual DMF, followed by drying in an oven. PVDF/ST powders with ST weight percentages of 2.5, 5.0, 7.5, and 10 wt% were prepared with different amounts of 0.1, 0.2, 0.3, and 0.4 g of ST. The corresponding powders were designated as ST_2.5, ST_5.0, ST_7.5, and ST_10, respectively. As a control group, 0.2 g of ST and 3.8 g of PVDF were dispersed in 10 ml of ethanol and 30 ml of ethanol, respectively. The ST-dispersed solution was added to PVDF-dispersed solution drop by drop. The completely dried powder was designated as dis_ST_5.0. Subsequently, all produced powders were processed by uniaxial hot-pressing method at 190 °C and 20 bar for 2 h.
We used the friction coefficient and specific wear rate as the tribological properties. Thus, we employed STM-Micro Tribology Tester (Hanmi Industry Ltd.) for a wear resistance test, as previously reported, which operated for recipro-mode [29]. To be specific, the distance of stroke for each cycle was 20 mm, the cycle frequency was 0.5 Hz, and the total distance of the test was 300 m for the friction coefficient. In addition, we use 900 m of the total distance to verify the specific wear rate. 3 N of the normal force was applied. Surface images of the composites after the wear tests were acquired via SEM.
The structures of powders and composites were investigated with field-emission scanning electron microscopy (FE-SEM, JSM-7800F, JEOL Ltd.) and transmission electron microscopy (TEM, JEM-F200, JEOL Ltd.). Functional groups of powders and composites were characterized by Fourier transform infrared spectroscopy (FT-IR, Spectrum 100, Perkin Elmer). The thermal properties of composites were analyzed with differential scanning calorimetry (DSC, TA Instrument, USA) from -50 to 200 °C at a heating rate of 10 °C min-1.
Scheme 1 briefly represents the processes for preparing PVDF/ST composites. In the preparation, DMF and ethanol were utilized as a solvent and a non-solvent for PVDF, respectively. A dissolution of PVDF in ST-dispersed DMF resulted in deep penetration of PVDF into the cavity inside SnO2 nanotubes, which enables contact over a large surface area between SnO2 nanotubes and polymer chains of PVDF. Good interfacial contact between PVDF and ST affects chain rearrangement resulting from mechanical and thermal deformation during hot pressing, which contributes to the high tribological performance of the composites. However, blending PVDF and ST in ethanol led to poor interfacial contact due to non-infiltration of PVDF into interior of ST. During the first mixing, when a small amount of PVDF is mixed with ST in DMF, uniformly mixed PVDF/ST powders prepared by infiltration of PVDF into ST can be obtained, which leads to excellent interfacial contact between them. As blending a large amount of PVDF with ST resulted in aggregation of ST particles and non-uniform mixing, a small amount of PVDF is utilized during the first mixing. In the second mixing, PVDF/ST was mixed with a large amount of PVDF, which provides the majority of PVDF/ST composites, followed by precipitation in ethanol as a non-solvent. Thus, uniform PVDF/ST composites in which ST is homogeneously dispersed and penetrated with PVDF can be prepared through two stages of the mixing process, as shown in Fig. 1.
Scheme 1. Schematic diagram of the fabrication procedure for PVDF/ST composites used in this study.
Fig. 1. Digital photographs of various PVDF/ST composites.
Hot pressing generally enhances the β- and γ-bands of PVDF due to the rearrangement of the polymer chains by mechanical and thermal deformation. Inorganic materials are known to suppress the formation of β- and γ-bands of PVDF during hot pressing. ST in PVDF/ST powders can effectively suppress the formation of β- and γ-bands in PVDF during hot pressing, promoting the formation of α-band. The α-band is advantageous for a high tribological property, while β- and γ-bands are disadvantageous for it. Thus, it is expected that ST will play an important role in the processes of manufacturing composites through hot pressing, improving the tribological performance of the composite.
The surface structure of ST obtained after calcination at 500 °C is shown in SEM images in Fig. 2(a) and (b), revealing uniformity in size and shape. The ST structure with uniform size and shape had a diameter scale of several hundred nanometers and a length scale of several micrometers. The specific surface area of ST, determined by the BET method, was 14.0 m2 g-1. The XRD diffraction pattern of ST in Fig. 2(c) revealed the highly crystalline property of the SnO2 rutile phase (JCPDS.41-1445), indicating the successful synthesis of SnO2 nanotubes. The surface of the PVDF/ST powders was also investigated by SEM analysis in Fig. 3. In Fig. 3(a) and (b) of dis_ST_5.0 prepared with ethanol, the morphologies of PVDF and SnO2 nanotubes were clearly observed. The original shape of the PVDF spheres was maintained owing to its non-dissolution in ethanol. Thus, there are large empty spaces between ST and PVDF, indicating a poor interfacial contact between them. In the preparation methods using DMF as a solvent for penetration of PVDF inside ST, the change in the morphology of PVDF is apparently shown in Fig. 3(c) and (d) in ST_2.5. Close contact between ST and PVDF was observed without voids between them, and dissolved PVDF penetrated inside the nanotube. The incorporation of more than a certain amount of ST into PVDF resulted in a similar morphology of PVDF/ST and good interfacial contact with penetration of ST into the nanotube. Thus, the results of surface SEM of PVDF/ST powders clearly confirmed that DMF dissolves PVDF and enables its penetration inside PVDF, thereby showing good interfacial contact.
Fig. 2. SEM images of ST obtained by electrospinning and calcination at low (a) and high (b) magnification and the corresponding XRD patterns (c).
Fig. 3. SEM surface images of PVDF/ST powders before hot pressing: (a, b) dis_ST_2.5; (c, d) ST_2.5; (e, f) ST_5.0; (g, h) ST_7.5; (i, j) ST_10.0.
PVDF/ST powders prepared by various methods were characterized using TEM analysis to compare the interfacial contact between PVDF and ST. In Fig. 4(a) and (b), TEM images of dis_ST_5.0, prepared in only ethanol, obviously revealed that undissolved PVDF separated from ST without penetration into nanotubes due to the insolubility of PVDF in ethanol. This confirmed that the preparation of dis_ST_5.0 powder led to poor interfacial contact between PVDF and ST. PVDF/ST powders prepared in DMF and ethanol showed that the PVDF spheres were dissolved and combined together. It also penetrated into the hollow inside the nanotubes, which resulted from its dissolution in ST-dispersed DMF. Thus, ST in the powder existed homogeneously inside the dissolved and combined PVDF structure, which resulted in a good interfacial contact between them. Structures with good interfacial contact between PVDF and ST were clearly observed, regardless of the number of ST incorporated. Therefore, the preparation methods for PVDF/ST powder resulted in a significant difference in the structure and interfacial contact.
Fig. 4. TEM images of PVDF/ST powders before hot pressing: (a, b) dis_ST_2.5; (c, d) ST_2.5; (e, f) ST_5.0; (g, h) ST_7.5; (i, j) ST_10.0.
Various powders before hot pressing and composites after hot pressing were investigated with FT-IR analysis in Fig. 5(a) and (b). Compared with PVDF powders, PVDF films obtained after hot pressing exhibited a strongly enhanced β-phase band at 841 cm-1 and a slightly enhanced α-phase bands at 975, 796, 763, and 614 cm-1. It was based on the mechanical and thermal deformation of polymer chains by hot pressing. It is well known that this change improves the tribological properties of the composite in that the α- and β-phase bands of PVDF enhance and deteriorate the tribological performance, respectively. Upon introduction of ST into composites, the intensification of β-bands was suppressed, while the formation of α-bands was remarkably promoted in all regions of the wavenumber. The enhancement of α-bands in PVDF/ST composites was observed regardless of the preparation methods and the amount of ST. Therefore, it confirmed that ST effectively induces a PVDF structure, which is favorable for the high tribological performance of the composites.
Fig. 5. FT-IR spectra of (a) blended PVDF/ST powders before the hot pressing process and (b) fabricated PVDF/ST composites after the hot pressing process, and DSC curves for validating, (c) the effect of ethanol dispersion in DMF dissolution method and (d) the effect of the amount of ST fillers in DMF dissolution method.
To further investigate the behavior of PVDF chains, an XRD apparatus was also included in this study (Fig. S1 in supplementary information). As we previously reported in Fig. 2(c), several facial arrangements of ST were observed in fabricated ST composites, such as (110), (101), (200), (211), (220), (310), (301), and (321) due to the existence of ST in composites. In the case of pure PVDF composite, both strong α-peaks and β-peaks were observed. However, as the ST content was increased, the α-peak of PVDF increased, and the β-peak of PVDF decreased simultaneously. Consequently, α(100) and α(020) of ST_10 composite have become sharper compared to pure PVDF and ST_5.0 composites. This selective increase in α-arrangement resulted from the interaction of PVDF chains and ST fillers. Due to the large surface area of ST, more PVDF chains can interact with ST and, ultimately, display a noticeable difference in peaks among composites. It agrees well with the results of previous FT-IR spectra and previous studies [[28],[29]].
Thermal properties of PVDF/ST composites were characterized using DSC analysis, with measurements performed at a heating rate of 10 °C/min, as shown in Fig. 5(c) and (d). DSC curves of various composites revealed the effect of interfacial contact between ST and PVDF on PVDF/ST composite and difference in preparation methods. DSC curves of bare PVDF and PVDF/ST composites in Fig. 5(c) reveal a comparison of the effect of added ST on PVDF/ST composites. Upon introduction of ST into PVDF, ST_5.0 and dis_ST_5.0 exhibit almost the same melting temperature (Tm) of 159.7 °C, which is lower than 161.1 °C of PVDF composite. This was a result of the enhancement of the α-phase band and the suppression of the β-band in the PVDF structure in the PVDF/ST composites. Fig. 5(d) shows the change with increasing number of ST in PVDF/ST composites. ST_2.5 exhibited Tm of 159.9 °C, which was higher than that of ST_5.0. Tm of ST_7.5 and ST_10 were 159.8 °C, which was lower than that of ST_5.0. The incorporation of a large number of ST into the PVDF/ST composites did not lead to a significant difference in the thermal behaviors of the composites. The result revealed that the addition of more than a certain amount of ST resulted in a sufficiently good interfacial contact between ST and PVDF.
The friction coefficient is analyzed in this section as a representative tribological performance (Fig. 6(a), Table 1). Initially, two factors are engaged in the friction coefficient. Metal oxides or metals have a high friction coefficient, which can be inferred from the fact that it is necessary to add oils during wear tests [31], but ST is able to interact with PVDF chains and leads to the α-phase arrangement at the same time (Fig. 5(b)). Therefore, ST_2.5 exhibited the best initial performance (0.101) in consideration of both two factors, and the excessive amount of ST leads to increased initial friction coefficient. It is critical for tribology to maintain low friction coefficient, so we tested the coefficient for 240 min. As the content of ST increased, the friction coefficient at 180 and 240 min decreased until the content reached 7.5 wt%. The result stems from the self-lubricating effect of the combination of PVDF and ST filler. However, the coefficient was again increased due to the nature of the high friction coefficient [31]. The specific wear rate shows how much the fabricated composite was scratched on the surface. The more the content of the ST filler increased, the less the specific wear rate was observed. The specific wear rate of ST_10 composite (0.412) is even lower than that of the pure PVDF composite (0.596), which results from the penetration of PVDF into the ST filler.
Fig. 6. (a) Friction coefficient of fabricated composites used in this work. The initial, 180 min, and 240 min values were employed, and an average value was used at each point, while 40 s measurements were carried out at 25.0 °C and 45.0% humidity; (b) specific wear rate of fabricated composites used in this work. Total weight loss after 12 h, 30 min of surface wearing was used to calculate the specific wear rate at 25.0 °C and 45.0% of humidity.
Table 1 Tribological performance of fabricated composites at 25 °C and 45% humidity (Friction coefficient values are average values of 40 s, and the specific wear rate was measured over 12 h 30 min).
| Samples | Friction coefficient | Specific wear rate (10-4 mm3 N-1 m-1) | ||
|---|---|---|---|---|
| 0 min | 180 min | 240 min | ||
| PVDF | 0.101 | 0.468 | 0.480 | 0.596 |
| ST_2.5 | 0.097 | 0.447 | 0.465 | 1.01 |
| ST_5.0 | 0.143 | 0.466 | 0.462 | 0.823 |
| ST_7.5 | 0.143 | 0.350 | 0.408 | 0.638 |
| ST_10 | 0.142 | 0.414 | 0.462 | 0.412 |
| dis_ST_5.0 | 0.120 | 0.441 | 0.446 | 0.926 |
ST_5.0 (0.143 at the initial point and 0.462 after 240 min) shows a slightly higher friction coefficient than dis_ST_5.0 (0.120 at the initial point and 0.446 after 240 min) at all times (Fig. 6(a), Table 1). This means that we obtained a better composite in terms of friction coefficient without the process of dissolving DMF. However, we should concentrate on the specific wear rate as well as the friction coefficient at the same time. The difference in specific wear rate between the two composites is 11% (0.823 for ST_5.0 and 0.926 for dis_ST_5.0), while that of friction coefficients is only 3.5%. Therefore, we concluded that the DMF dissolution method is necessary for enhancement of overall tribological performance, because the high specific wear rate of PVDF is a weak point in terms of ingredients of tribology. The worn surface of the tested composites was examined using surface SEM images (Fig. S2). The pure PVDF and dis_ST_5.0 composites display rougher surface debris compared to ST_5.0 and ST_10. The result is consistent with the specific wear rate (Fig. 6(b)). In addition, ST filler is also found on the worn surface (Fig. S2(d) and (h)).
We represented several examples to enhance the tribological properties with inorganic fillers in PVDF matrix (Table 2) such as reduced graphene oxide, carbon nanofibers, and carbon nanotube [[21],[24],[32]]. Also, TiO2 nanoparticles and MgO nanosheets were also employed [[28],[29]] for tribological applications. It should be noted that the recipro-mode was utilized in our study to replicate the circumstance of earthquake unlike other studies, and thus the initial friction coefficient was used for a fair comparison. Most of the studies were performed less than 1 h except for our previous study [29] with MgO nanosheet and this study. The PVDF-filled ST composites exhibited excellent performance, which is similar to the PVDF/MgO nanosheet composite. It demonstrates the pivotal role of ST as a self-lubricating material due to a large interactive area and PVDF chain rearrangement.
Table 2 Recent research on the tribological properties of PVDF/filler composites.
| Filler | Composite fabrication | Friction coefficient | Measurement time (min) | Ref. |
|---|---|---|---|---|
| Epoxy/Carbon nanotube | Solution-mixing | 0.45 | 40 | [21] |
| Polyimide/Carbon nanofiber | Wet ball milling/Hot-press | 0.19 | 30 | [24] |
| TiO2 nanoparticles | Dry-mixing/Uniaxial hot-press | 0.17 (Kinetic) | 50 | [28] |
| Mesoporous MgO nanosheet | Solution-mixing/Uniaxial hot-press | 0.091 | 250 | [29] |
| Reduced graphene oxide | Solution-Mixing | 0.1 | 60 | [32] |
| SnO2 nanotube | Solution-Mixing | 0.097 | 240 | This study |
A novel class of PVDF/ST composites was fabricated to enhance tribological properties. In particular, we used electrospinning and the sintering process to produce ST. The produced ST was dispersed in a PVDF/DMF solution to penetrate the PVDF chains into the ST and subsequently precipitated in ethanol. The effect of the impregnation method was also investigated in this study. We observed well-penetrating PVDF/ST mixtures in TEM images. In addition, uniaxial hot pressing was employed to fabricate composites that we used in this study at 190 ºC and 20 bar for 2 h. We observed well-dispersed, PVDF-penetrated ST particles in the cross-sectional images of fabricated composites. Owing to the large surface area of the ST filler, the selective arrangement of PVDF chains (α-phase) was detected in FT-IR spectra and XRD patterns. A well-arranged PVDF chain from the interaction between the PVDF chain and the ST filler resulted in the enhancement of tribological properties. A recipro-type module was utilized to simulate an earthquake for each test. The friction coefficient of PVDF was significantly enhanced in terms of long-term stability, i.e. from 0.480 for PVDF to 0.408 for ST_7.5. Moreover, the specific wear rate was consistently improved from 2.5 to 10 wt% of ST fillers, which ranged from 1.01 to 0.412. This noticeable enhancement in tribological performance could be a breakthrough in earthquake-proof building construction.
This work was supported financially by a grant from the National Research Foundation (NRF) of South Korea, funded by the Ministry of Science, ICT and Future Planning (Nos. NRF-2017R1A4A1014569 and NRF-2018M3A7B4071535).
Supplementary material related to this article can be found, inthe online version, at doi: https://doi.org/10.1016/j.jmst.2019.07.038.
WeChat
/
| 〈 |
|
〉 |
