Skip to content
BY-NC-ND 3.0 license Open Access Published by De Gruyter May 27, 2013

Densification treatment and properties of carbon fiber reinforced contact strip

  • Hua Yuan , Chengguo Wang EMAIL logo , Shan Zhang , Xue Lin and Meijie Yu

Abstract

The high temperature caused by current-carrying wear could affect the thermal reliability of resin-based contact strip greatly. This study adopted liquid-phase impregnation-carbonization (IC) technique to improve the thermal stability and densification of carbon fiber reinforced contact strip (CFRCS). The influence of this method was investigated by scanning electron microscopy, Fourier transform infrared spectrometry, thermal gravimetric analysis and energy-dispersive spectrometry; meanwhile, specimen composition and friction and mechanism properties were also analyzed. The results show that heat treatment is helpful in improving the material’s temperature tolerance. When specimens undergo IC treatment four times, resistivity and wear rate would reduce gradually under impregnating conditions of carbonization temperature (800°C), dipping liquid concentration (60%), and dipping temperature (60°C). IC treatment is effective in reducing material porosity and improving the impact resistance performance compared with only carbonized sample. Densification treatment can also improve the samples’ compressive strength and bending strength. The main wear mechanisms of CFRCS-25 and CFRCS-800 against copper with electrical current are similar; these are arc erosion wear and oxidation wear accompanied by adhesive wear. Adhesive wear and oxidative wear is more severe for CFRCS-25 than CFRCS-800.

1 Introduction

In recent technological developments in electric railways, major efforts have been devoted increasing the speed of trains. For example, special tribological pairs of the electric railway contact strip against copper contact wire were made [1, 2]. So far, the contact strips used in electric locomotives included copper-based powder metallurgy slider, steeped metal carbon slider, and carbon contact strip; however, their application has been limited by their inferior mechanical properties [3]. Carbon fiber reinforced composites are primarily developed and designed for structural applications due to the fibers’ excellent properties such as high specific strength and modulus [4–6]. Thus, this study prepared carbon fiber reinforced composite as contact strip. This material uses phenolic resin as binder and carbon fiber as reinforced phase.

Phenolic resins are considered to be the first polymeric products produced commercially from simple compounds of low molecular weight [7, 8]. Phenolic resins continue to be used for a wide variety of applications, such as molding powders, laminating resins, adhesives, binders, surface coatings, and impregnants with low cost, versatility, heat and flame resistance, durability, strength and stiffness, low toxicity, and ease of processing. However, as important current-carrying components, contact strip surface temperature will rise seriously, caused by friction heat and arc ablation [9, 10], and in the aerobic environment of 300°C, phenolic resin will start to be dissociated, which will shorten the service life of the resin matrix contact strip [11–13]. Therefore, heat treatment and densification processing is needed to avoid contact strip performance decline brought about by resin cracking.

Generally, the composite densification technology includes liquid-phase impregnation method, chemical vapor deposition, and chemical vapor infiltration [14]. The liquid-phase impregnation method is simple and easy to operate and implement [15]. Therefore, this study uses liquid-phase impregnation method to achieve contact strip material densification.

In this paper, we focus on researching the densification and properties of carbon fiber reinforced contact strip (CFRCS). The characteristic and morphological changes of contact strip were investigated by scanning electron microscopy (SEM), thermal gravimetric (TG) analysis, and Fourier transform infrared (FTIR) spectroscopy; meanwhile, the samples’ friction performance and mechanism were researched.

2 Experimental

2.1 Specimen preparation

The preparation process of CFRCS is shown in Figure 1. The chemical composition of CFRCS used in this paper is as follows: Cu mesh, 15 wt%; carbon fiber, 22 wt%; phenolic resin, 33 wt%; graphite, 22 wt%; Cu powder, 5 wt%, and nitrile rubber, 5 wt%. The specimens are made by compression molding at a pressure of 100 MPa and temperature of 170°C.

Figure 1 Scheme of preparation process of CFRCS.
Figure 1

Scheme of preparation process of CFRCS.

2.2 Impregnation process

Figure 2 shows a schematic diagram of the impregnating device. The impregnating process is as follows: First, dried samples are placed into an impregnating tank and the impregnating tank is sealed. The pumped impregnating tank is vacuumed with a suction pump to achieve a negative pressure of about -0.09 MPa when the equipment temperature reaches the specified value and held for 5 min. Then the valve of the storage tank is opened and heated phenolic resin ethanol solution is pumped into the impregnating tank. Nitrogen gas is added to increase the pressure up to 1 MPa and held for 2 h.

Figure 2 Impregnating device schematic diagram.
Figure 2

Impregnating device schematic diagram.

2.3 Test apparatus

Resistivity of the composites was measured with a four-probe electrical technique, manufactured by Shandong University, China. The resistivity (ρ) of samples is obtained by Eq. (1):

where U is the applied voltage and I is the measured alternating current. S and Lu are the cross-sectional area and length of the specimen, respectively.

A pendulum impact testing device was used to test the impact strength of the specimen. The processing parameters were as follows: temperature T=15–35°C, horizontal velocity ν=2.9 m/s. The impact strength a (kJ·m-2) is given by Eq. (2):

where Estands for impact energy (J), b stands for specimen size in length (mm), and d stands for specimen size in width (mm).

The contact strips’ porosity was tested using Archimedes drainage method. The porosity (%) is given by Eq. (3):

where m1 (g) is the weight of the dry contact strip samples. The samples were placed in water and boiled for 2 h, and then the samples’ surface water was gently wiped off using a wet cloth. m2 is the samples’ weight with water in the pores. m0 is the weight of copper wire. The sample was fastened with copper wire and the contact strip sample was immersed in water completely. At this time, the weight of sample was m3.

Friction and wear tests (shown in Figure 3) were conducted using an HST-100 high-speed electrical wear tester, made by Henan University of Science and Technology, China, with a copper disk and pantograph contact strip as tribological pairs. The rotational speed was 40 m/s and the size of a single friction material specimen was 12 mm×9 mm×25 mm. The friction pair’s loading was 70 N, the current density was 80 A·cm-2, and the process was held for 30 s. All the specimens were ground with 600-grid paper to have a uniform standard surface, as the surface finish of the specimens will influence the friction and wear characteristics. The weight of each specimen was tested before and after the friction test. Friction force was continuously recorded during the test. The volume wear rate is given by Eq. (4):

Figure 3 Schematic diagram of wear testing equipment.
Figure 3

Schematic diagram of wear testing equipment.

where Ws stands for sample volume wear rate (mm3·N-1·m-1); ΔV is wear volume (mm3) and ΔW is wear weight (g); N is load (N); L is friction distance (m); ρ stands for sample density (g·cm-3); ν is grinding wheel speed (m·s-7); and t is friction time (s).

The wear and friction tests of composites were analyzed by a scanning electron microscope (SEM US-70) operated at 15.0 kV.

TG and differential thermal analysis (DTA) used a simultaneous thermal analyzer. FTIR spectrum was measured in a Bruker Alpha-type infrared spectrometer using KBr compression method: scan range, 400–4000 cm-1; scanned 64 times; resolution of 4 cm-1.

3 Results and discussion

3.1 Analysis of phenolic resin structure and thermal stability

Figure 4 shows the FTIR spectrogram of phenolic resin at different temperatures. Phenolic resin undergoes pyrolysis and solidification when heated to 800°C in nitrogen gas. During pyrolysis, several thermal chemical reactions occur such as condensation of neighboring rings accompanied by the formation of methylene bridges; decomposition of certain groups, and chain scissions with release of gaseous by-products such as water, carbon monoxide, carbon dioxide, methane, hydrogen, etc. [5, 16]. From the phenolic resin infrared spectrum after treatment at 600°C, it can be seen that heat treatment makes -OH stretching vibration absorption peak of nuclear phenol at 3300 cm-1 and -CH2- symmetric stretching vibration at 2924 and 2896 cm-1. The methylene plane shear vibration peak at 1470 cm-1 gradually weakened and retained the peak of the aromatic ring. This shows that the phenolic resin treatment at 600°C resulted in intermolecular dehydration, cyclization, and intramolecular dehydrogenation, and the methylene bond that connects benzene fractured and formed polyacenes (a schematic diagram of the phenolic resin pyrolysis in inert gas is shown in Figure 5). At a heat treatment temperature of up to 800°C, phenolic resin will generate the six-party crystal structure. From the infrared spectrum, it can be observed that all characteristic peaks are not obvious. The matrix internal carbon atom changed to sp2 hybridization from sp3 hybridization; the conductive performance of the carbon matrix will be improved obviously and form many through-holes in the matrix by decomposition gas escaping. Consequently, the phenolic resin characteristic peak disappeared with treatment temperature rising from 25°C to 800°C. Some methylene bonds broke because of high activation. Condensation reaction occurred between phenolic hydroxy groups (-OH); structural decomposition occurred, and the phenol-formaldehyde structure was eventually transformed into an amorphous carbon structure.

Figure 4 FTIR spectrogram of phenolic resin at different temperatures.
Figure 4

FTIR spectrogram of phenolic resin at different temperatures.

Figure 5 The phenolic resin pyrolysis schematic diagram in inert gas.
Figure 5

The phenolic resin pyrolysis schematic diagram in inert gas.

Typical TG and DTA curves in air of CFRCS without treatment (CFRCS-25) and after treatment at 800°C (CFRCS-800) are given in Figure 6. It can be seen that the initial reaction temperature of CFRCS-800 is higher than that without treatment. When weight loss is about 10%, the temperatures of CFRCS-25 and CFRCS-800 are 392°C and 465°C, respectively. After two thermal weight loss stages, the residual mass rate of CFRCS-800 is higher than that of CFRCS-25. Based on the above analysis, a conclusion can be drawn that heat treatment is helpful to improve CFRCS temperature tolerance.

Figure 6 TG and DTA curves of CFRCS before and after treatment.
Figure 6

TG and DTA curves of CFRCS before and after treatment.

However, the density of CFRCS-800 will decline dramatically as a result of phenolic resin pyrolysis and many holes will generate. Therefore, specimen densification is needed.

3.2 Impregnated liquid concentration and impregnation temperature

Dipping liquid viscosity and temperature are key factors for impregnation effect. Therefore, a rotary viscometer was used to study the changes in impregnating liquid viscosity with concentration and temperature, and the results are shown in Figure 7. It is well known that viscosity increases as the solution concentration increases, whereas it decreases with increasing temperature. The relationship between liquid viscosity and temperature is shown in Eq. (5). The polymer internal free volume enlarges with increasing temperature; molecular chain transition ability heightens and liquidity becomes better, so the viscosity drops. Considering the solubility and viscosity of phenolic resin on impregnation effect, the experimental impregnating conditions chosen were dipping liquid concentration of 60% and dipping temperature of 60°C.

Figure 7 Relationship between impregnation solution viscosity and temperature at different concentrations.
Figure 7

Relationship between impregnation solution viscosity and temperature at different concentrations.

where η is viscosity, ΔEη is flow-activation energy, A and R are constant, and T is solution temperature.

3.3 Effect of impregnation-carbonization time on resistivity and abrasion

When specimens undergo a one-time liquid impregnation process, the composite products’ structure is still loose and its internal structure contains many pores. However, the existence of holes has a great influence on the material properties. Therefore, repeated impregnation-carbonization (IC) process is needed to gradually fill in the pores to achieve the requirements of density and performance. Here, the calcination times on the wear rate and resistivity of contact strip was studied and presented in Figure 8 (0 is no processing sample performance; 1–4 is the performance of the sample carbonized one to four times). The data in the figure means the change in amplitude of the sample performance after one-time IC.

Figure 8 Resistivity and wear rate with the time of carbonization-impregnation.
Figure 8

Resistivity and wear rate with the time of carbonization-impregnation.

From Figure 8, it can be seen explicitly that resistivity and wear rate reduce gradually along with the increase of the IC times. Contact strip specimen resistivity decreases significantly (about 50% after one-time carbonization). Phenolic resin cracks after heat treatment and becomes amorphous carbon, which has a loose structure [residual carbon rate is about 50% (Figure 9)]. Carbon skeleton resistivity produced by phenolic resin thermal cracking is lower than phenolic resin resistivity [17]; at the same time, copper relative content increases as material losing weight, so the resistivity obviously drops. Densification processing is helpful to increase the density of material and reduce holes; therefore, the material conductive performance is stable. Friction pair microscopic contact area is larger in the current-carrying wear process, which can effectively reduce arc ablation caused offline. In addition, the fiber does not fall off from the sample easily because it is closely coated in the matrix, and wear rate decreases with the increase in the IC times. Moreover, with the increase in IC times, sample performance change in amplitude becomes smaller. Thus, in order to increase the impregnating effect and reduce cost, sample IC times should be four times.

Figure 9 TG and DTA curve of phenolic resin in nitrogen after curing.
Figure 9

TG and DTA curve of phenolic resin in nitrogen after curing.

3.4 Effect of calcinations time on compressive strength and bonding strength

Compression stress and strain relation of different samples is shown in Figure 10. The data of compressive strength is calculated according to Eq. (6).

Figure 10 The compression load-displacement curves of different samples.
Figure 10

The compression load-displacement curves of different samples.

where σy is the compressive strength (MPa), F is the compression maximum load (N), and S is the sample sectional area (mm2).

Analyzing the compressive strength data, it can be concluded that the stress-strain curves do not show pseudoplastic behavior before the specimen failure and have the obvious linear features. According to the compression strength hybrid model [Eqs. (7) and (8)], the compressive strength of composite is mainly affected by carbon fiber strength and resin matrix strength. For the no-roasting sample, curing resin has a compact structure, carbon fiber is destroyed first under external load, and the sample compression strength value is bigger. However, the amorphous carbon structure of the roasting sample is loose; its compressive strength fell about 44.9% more than that of the untreated sample. The matrix nonlinear shear performance plays a main failure effect, so the curves in Figure 10 (1, 2, and 3) show a certain pseudoplastic behavior. In addition, the densification treatment can effectively improve the fiber and resin interface combination and reduce the porosity of the matrix; the matrix bearing capacity is improved, so the compressive strength improves gradually.

Fiber-first destruction:

Matrix first destruction:

where σy is composite compressive strength (MPa); σCF and σM stand for carbon fiber longitudinal compression strength and resin longitudinal compression strength, respectively (MPa); VM, VCF is the volume of matrix content and carbon fiber, respectively (%); EM, ECF is matrix and carbon fiber elasticity modulus, respectively.

Figure 11 shows the bending stress-strain curves of samples. Bending strength is calculated according to Eq. (9). From the graph, stress and strain keep a basic linear correlation before the load up to maximum value, and after the peak, stress presents stepped down, this is a typical “false elastic-plasticity effect”. Composite bending damage is mainly due to matrix damage, fiber bundle axial damage, and fiber beam shear damage. The sample matrix undergoes stress concentration and forms microcracks first with the bending load; microcracks expand along the direction of the interface and converge with other microcracks and form a macroscopic crack; under the critical stress, the macroscopic crack will be extended, which eventually leads to the sample being damaged. When the sample is heat treated, the resin interior would form some structural defects because of cracking gas effusion, and the resin matrix crack expansion is rapid; at the same time, the resin and fiber interface bonding force is weak, and the sample is destroyed by fiber prior to pulling out, so bending strength is relatively low. During the densification process, the resin has been filling the interstices, and the structure gradually becomes denser. The bond between fiber and resin increases, and the fiber can effectively transfer load as a secondary bearing surface and slow crack growth, so sample bending strength increases.

Figure 11 The flexural load-displacement curves of different samples.
Figure 11

The flexural load-displacement curves of different samples.

where σz is bending strength (MPa), F is broken load (N), Ls is supporting point distance (mm), W is sample width (mm), and b is sample thickness (mm).

3.5 Specimen surface topography changes before and after impregnation

Figure 12 shows the specimen’s surface topography before and after impregnation, which is obtained from the impact section. Before heat treatment (Figure 12A), the resin structure is compact. Fiber fracture is stepped in favor of transferring stress; therefore, specimen impact resistance performance is good. After heat treatment at 800°C (Figure 12B), there are many pores with different apertures in the specimen interior, and under impact pores will exist between the fiber and matrix. The interface bonding strength is relatively lower at this moment. From Figure 12C, pores become decrescent after IC treatment four times. In addition, the porosity values (P) of three samples measured by Archimedes drainage method are 1.84%, 15.21%, and 7.12%, respectively. Resin dipping into the sample will become the filling, gain weight and form part of the closed hole in the carbonization process. Through four IC treatments, opening porosity is down about 50%. A small amount of porosity is beneficial to improve the properties of materials [18]. Although composite material is always filled with small holes after liquid-phase IC process, Ju et al. [19] propose that appropriate pores existing in the composite are helpful for formation of friction film, which has a lubrication effect, which is beneficial to improve the wear resistance.

Figure 12 SEM photo of specimens’ surface topography before and after impregnation. (A) Untreated; (B) after one time carbonization; (C) after four times IC; (D) impact strength.
Figure 12

SEM photo of specimens’ surface topography before and after impregnation. (A) Untreated; (B) after one time carbonization; (C) after four times IC; (D) impact strength.

At the same time, samples’ impact performance was tested, which is shown in Figure 12D. Impact strength increases with increasing IC times, improving about 32.1% after four times IC treatment than that of just one-time carbonization sample.

On the basis of the above observation, it can be clearly seen that liquid-phase IC treatment is effective to reduce material porosity and improve the impact resistance performance compared to the carbonized sample.

3.6 Friction and wear mechanism

Contact strip life depends on quality loss in the process of sliding directly, so the study of friction and wear is very necessary. Figure 13 shows the SEM micrographs of the worn surface of different contact strips. Grooves can be seen in Figure 13A and C, which are characteristic of an adhesive wear mechanism. Cracks and arc craters can be found on the surface of CFRCS (Figure 13A and C). However, the wear debris of untreated contact strip appears to be solid spheres with conglomerate appearance. Upon sliding with current, electric spark is induced between tribological pairs [20, 21]. Arc discharge and arc heat will increase the temperature of the worn surface. However, resin matrix without heat treatment is thermally decomposed, much small-molecule volatile substance is produced, and matrix structures will become incompact, leading to carbon fiber being easily pulled out (Figure 13B) and wear rate increase. Friction stability will be poor. Worn surface of CFRCS after four IC treatments appears compact, and the surface groove is shallow, which is engendered by wear with current.

Figure 13 SEM graphics for specimen. (A) and (B) SEM photo untreated; (C) and (D) SEM photos after treatment at 800°C.
Figure 13

SEM graphics for specimen. (A) and (B) SEM photo untreated; (C) and (D) SEM photos after treatment at 800°C.

The data of worn surface elements analyzed by energy-dispersive spectrometry are listed in Table 1 (percentage by weight). The copper content of wear debris of CFRCS-25 is higher than that of CFRCS-800. The main reason is that the copper content of CFRCS is 15 wt% and transfer of material from disk to contact strip during electrical sliding occurs. This confirms further that adhesive wear exists and the wear to copper wheel is more severe for CFRCS-25 than for CFRCS-800. In addition, a large number of small spherical droplets are observed on the worn surface (Figure 13A). Those spherical droplets are <10 μm in diameter and are supposed to be caused by arc melting under current of 80 A. Meanwhile, the O/C content of CFRCS-25 is a little higher than that of CFRCS-800. This is because the oxidation by arc discharge is related to the arc erosion; the higher the degree of oxidation on the worn surface, the higher the degree of oxidative wear. Therefore, the oxidative wear is more severe for CFRCS-25 than CFRCS-800.

Table 1

Element composition on worn surface.

SpecimenComposition (%)O/C
COCu
CFRCS-2559.969.0932.950.1516
CFRCS-80078.428.4613.120.1079

Based on the above analysis, the dominant wear mechanism of CFRCS-25 and CFRCS-800 can be given as follows: the main wear mechanisms of CFRCS-25 and CFRCS-800 against copper with electrical current are similar; these are arc erosion wear and oxidation wear, accompanied by adhesive wear. Adhesive wear and oxidative wear is more severe for CFRCS-25 than CFRCS-800.

3.7 Performance comparison of different contact strips

The CFRCS-25, CFRCS-800, and pure carbon contact strips’ performances were compared, and the results are shown in Table 2. Test temperature was room temperature, current density was 80 A/cm2, sliding velocity was 40 m/s, the load was 70 N, and wear time was 30 s.

Table 2

Different contact strips performance comparison.

CFRCS-25CFRCS-800Pure carbon contact strip
Density (g·cm-1)1.811.681.76
Resistivity (μΩ·m)8.634.535
Impact strength (kJ·m-1)7811.51.5
Wear rate (×105 mm3·N-1·m-1)4.35.43.7
Friction coefficient1.641.661.62
Flexural strength (MPa)16812344
Compressive strength (MPa)≥200160106

It can be seen from Table 2 that the mechanical properties of CFRCS-25 and CFRCS-800 (impact strength, bending strength, compressive strength) are better than that of pure carbon contact strip; the resistivity of CFRCS-25 is smaller. Under the same conditions, the wear rate of CFRCS-25 and the wear rate of pure carbon slide are similar, but resin CFRCS-25’s friction coefficient is high.

Thus, CFRCS-25 and CFRCS-800 materials’ comprehensive performance is excellent and is suitable for electric locomotives.

4 Conclusions

In order to improve the carbon fiber reinforced resin-based contact strip wear stability with current, heat treatment is necessary. FTIR and TG analysis results prove that heat treatment is helpful to improve the material’s temperature tolerance. When specimens undergo IC treatment four times, resistivity and wear rate reduce gradually in impregnating conditions at carbonization temperature of 800°C, dipping liquid concentration of 60%, and dipping temperature of 60°C. Impregnation treatment is effective to reduce material porosity and improve the impact resistance performance, compressive strength, and bonding strength compared to carbonized sample. The main wear mechanisms of CFRCS-25 and CFRCS-800 against copper with electrical current are similar; these are arc erosion wear and oxidation wear, accompanied by adhesive wear. Adhesive wear and oxidative wear is more severe for CFRCS-25 than CFRCS-800. CFRCS-25 and CFRCS-800 materials’ comprehensive performance is excellent and is suitable to apply in electric locomotive.


Corresponding author: Chengguo Wang, Carbon Fibre Engineering Research Center, Faculty of Materials Science, Shandong University, Jinan 250061, China, e-mail:

This project was supported by the Major State Basic Research Development Program of China (973 Program) (Grant No. 2011CB605601), the National Natural Science Foundation of China (Grant No. 50902088), and the National Natural Science Foundation of Shandong Province (Grant No. 2009ZRB019C8).

References

[1] Javier G, Rafael G, Gabriel G. J. Transp. Geogr. 1996, 4, 227–238.Search in Google Scholar

[2] Shangguan QQ, Cheng XH. Appl. Surf. Sci. 2007, 253, 9000–9006.Search in Google Scholar

[3] Fröidh O. J. Transp. Geogr. 2008, 4, 268–277.Search in Google Scholar

[4] Karnika SR, Gaitondeb VN, Campos Rubioc J, Esteves A. Mater. Des. 2008, 29, 1768–1776.Search in Google Scholar

[5] Boudeau N, Liksonov D, Barriere T, Maslov L. Mater. Des. 2012, 40, 148–156.Search in Google Scholar

[6] Li W, Chen ZH, Li J, Chen XH, Xuan H, Wang XY. Mater. Sci. Eng. A. 2008, 485, 481–486.Search in Google Scholar

[7] Mutlu I, Oner C, Findik F. Mater. Des. 2007, 28, 480–487.Search in Google Scholar

[8] Jiang HY, Wang JG, Wu SQ. J. Funct. Mater. 2008, 6, 1024–1026.Search in Google Scholar

[9] Zhao H, Barber GC, Liu J. Wear 2001, 249, 409–414.10.1016/S0043-1648(01)00545-2Search in Google Scholar

[10] Azevedo CRF, Sinatora A. Eng. Failure. Anal. 2004, 11, 829–841.Search in Google Scholar

[11] Krstkova M, Filipb P, Weissa Z, Peter R. Polym. Degrad. Stab. 2004, 1, 49–60.Search in Google Scholar

[12] Ho SC, Lin JHC, Ju CP. Carbon 2005, 3, 491–502.10.1016/j.carbon.2004.09.028Search in Google Scholar

[13] Kim YJ, Kim M, Yun CH, Chang JY, Park CR, Inagaki M. J. Colloid. Interface. Sci. 2004, 274, 555–562.Search in Google Scholar

[14] Ning X, Ishida H. J. Polym. Sci., Part B: Polym. Phys. 1994, 32, 1121–1129.Search in Google Scholar

[15] Michoud V, Mortensen A. Composites 2001, 32, 981–996.10.1016/S1359-835X(01)00015-XSearch in Google Scholar

[16] Manocha LM, Bhatt H, Manocha SM. Carbon 1996, 7, 841–849.10.1016/0008-6223(95)00201-4Search in Google Scholar

[17] William DD. In Adhesion Science Engineering. New York, 2002, Chap. 20, pp. 869–945.Search in Google Scholar

[18] Li HJ, Fei J, Qi LH, Fu YW. J. Inorg. Mater. 2007, 6, 1159–1164.Search in Google Scholar

[19] Ju CP, Lee KJ, Wu HD, Chen CL. Carbon 1994, 5, 971–977.10.1016/0008-6223(94)90057-4Search in Google Scholar

[20] Cho YH. J. Sound. Vib. 2008, 315, 433–454.Search in Google Scholar

[21] Jia SG, Liu P, Ren FZ, Tianb BH, Zhenga MS, Zhoua GS. Mater. Sci. Eng. A. 2005, 398, 262–267.Search in Google Scholar

Received: 2012-12-19
Accepted: 2013-4-6
Published Online: 2013-05-27
Published in Print: 2014-01-01

©2014 by Walter de Gruyter Berlin Boston

This article is distributed under the terms of the Creative Commons Attribution Non-Commercial License, which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

Downloaded on 29.3.2024 from https://www.degruyter.com/document/doi/10.1515/secm-2012-0177/html
Scroll to top button