Experimental study on abrasive waterjet polishing for hard–brittle materials

Experimental study on abrasive waterjet polishing for hard–brittle materials

ARTICLE IN PRESS International Journal of Machine Tools & Manufacture 49 (2009) 569–578 Contents lists available at ScienceDirect International Jour...

2MB Sizes 0 Downloads 14 Views

ARTICLE IN PRESS International Journal of Machine Tools & Manufacture 49 (2009) 569–578

Contents lists available at ScienceDirect

International Journal of Machine Tools & Manufacture journal homepage: www.elsevier.com/locate/ijmactool

Experimental study on abrasive waterjet polishing for hard–brittle materials H.T. Zhu a, C.Z. Huang a,, J. Wang b, Q.L. Li a, C.L. Che a a b

Center for Advanced Jet Engineering Technologies (CaJET), School of Mechanical Engineering, Shandong University, Jinan 250061, China School of Mechanical and Manufacturing Engineering, The University of New South Wales, Sydney, NSW 2052, Australia

a r t i c l e in fo

abstract

Article history: Received 12 September 2008 Received in revised form 15 February 2009 Accepted 17 February 2009 Available online 27 February 2009

The rapid growth of hard–brittle materials necessitates the development of compatible machining techniques, especially for the precision machining. The abrasive waterjet (AWJ) machining is a powerful tool in processing hard–brittle materials. In the last decades, some of AWJ machining technologies, such as AWJ cutting, AWJ milling and AWJ drilling have gradually become mature and steady. However, a few investigations on precision surface machining for hard–brittle materials by AWJ had been carried out. In this research, the ductile erosion mechanism of hard–brittle materials by AWJ in small erosion angle has been analyzed. In theory, the ductile erosion can achieve micromaterial removal and the surface eroded is smooth and without any fracture. Based on the ductile erosion mechanism, the feasibility of polishing for hard–brittle materials by the AWJ has been investigated. A group of polishing experiments is performed. The polished surfaces of workpieces were observed with scanning electron microscope (SEM) and measured by atomic force microscopy (AFM). The results of these polishing experiments indicate that AWJ has a great potential to be used as a precision surface machining technology. & 2009 Elsevier Ltd. All rights reserved.

Keywords: Abrasive waterjet machining Hard–brittle material Ductile erosion Polishing

1. Introduction Hard–brittle materials, such as ceramics and glass, have become significantly important engineering material for numerous wear resistant and optical applications. Some hard–brittle material components need very high surface quality and free damage. At present, grinding, lapping and polishing are conventional processes for shaping hard–brittle materials in precision and ultraprecision machining. However, some shortcomings like low efficiency, low profile accuracy of machined surface and machining damages due to effects of machining force or heat are inevitable. The rapid growth of application of hard–brittle materials necessitates the development of compatible machining techniques. The precursor to abrasive waterjet (AWJ), modern waterjet technology, was initiated by Dr. Norman C. Franz in 1968, who was awarded with the first patent for a high-pressure waterjet cutting system. Based on this idea, the first commercial waterjet cutting system was developed to cut laminated paper tubes in 1971. In 1980, Dr. Hashish invented the process of adding abrasives to the plain waterjet and the abrasive waterjet was used for the first time to cut steel, glass and concrete. In 1983, the idea of entraining abrasive into the water stream was immediately followed by the  Corresponding author. Tel./fax: +86 531 88396913.

E-mail addresses: [email protected] (H.T. Zhu), [email protected] (C.Z. Huang), [email protected] (J. Wang), [email protected] (Q.L. Li), [email protected] (C.L. Che). 0890-6955/$ - see front matter & 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.ijmachtools.2009.02.005

development of the first commercial abrasive waterjet system for cutting automotive glass. The abrasive waterjet machining, which is based on the principle of rapid erosion by high-speed abrasive waterjet combined with rapid cooling by the water jet, is a powerful tool in processing various materials [1]. This technology can achieve faster machining speeds and leave a fine surface quality free of thermal distortion [2,3]. The AWJ machining has numerous potential applications and has been widely used for machining hard–brittle materials such as ceramic, glass and so on. The AWJ cutting is the earliest process applied widely in industry. However, the major obstructions that limit its application are the depth of cut and kerf quality [4]. A great deal of research has been done to improve the cutting performance and enhance the cutting capacity of AWJ cutting technology, including studies of the mechanism of the AWJ cutting process [5–7] and modeling for process control and optimization [8–11]. Some newly developed techniques such as cutting by the forward angling of the jet in the cutting plane, multipass cutting and controlled nozzle oscillation have been found to be effective in improving the cutting performance without additional costs to the machining process [12–14]. The AWJ milling is used in surface machining for difficult-to-machine materials like ceramics, composites and so on. The material removal mechanism, the machined surface characteristics and the effect of process parameters on machined surface quality have been studied [15–18]. The AWJ drilling is a process used frequently in industry by which small-diameter holes can be machined expediently. It is

ARTICLE IN PRESS 570

H.T. Zhu et al. / International Journal of Machine Tools & Manufacture 49 (2009) 569–578

Nomenclature

l

Fn Ft R

s11, s22, s33 principal normal stresses, Pa s13, s12,s23 principal shear stresses, Pa srr, syy, szz, srz, sry ,szy stress components, Pa sij general form of stresses, Pa

a y F u

vertical load, N horizontal load, N polar coordinate, m erosion angle, degree polar coordinate, m polar coordinate, m poisson’s ratio of material, dimensionless

critically important to control the size and shape of the hole and the quality of surfaces in drilling process by AWJ. Some studies have revealed some of the drilling mechanisms [19,20]. By controlling the jet’s pressure-time profile and the abrasive flow rate, it is proved that the hole of high quality can be drilled by AWJ [21,22]. In the last decades, much attention has been paid to the improvement of the AWJ machining technologies. Some of these, such as AWJ cutting, AWJ milling and AWJ drilling, have gradually become mature and steady. However, a few investigations on precision surface machining for hard–brittle materials by the AWJ had been reported. In this research, the ductile erosion mechanism of hard–brittle materials by AWJ in small erosion angle has been analyzed. Based on the ductile erosion mechanism, the feasibility of polishing hard–brittle materials by the AWJ has been investigated. Results of polishing experiments indicate that the AWJ has a great potential to be used as a precision surface machining technology.

2. Ductile erosion by AWJ in a small erosion angle The erosion process in high-pressure AWJ machining mainly includes impacting of solider particles and water jet. In the erosion caused by impacting of single solider particle, there are four kinds of mechanisms such as cutting, fatigue, melting and brittle fracture [23]. Clearly, these mechanisms generally do not act separately, but in combination. In the erosion process by water jet, there are two major actions, namely impacting by waterjet directly and shearing by lateral flow of waterjet. Under low pressure and small erosion angle, the impacting process of solider particles is dominant. Therefore, the material removal is caused mainly by the erosion of the solider particle in AWJ machining especially for hard–brittle material. At present, the erosion of a hard–brittle material when impacted by a high-speed abrasive particle is not understood in detail. However, if the dynamic effect is ignored and if it is assumed that the material response is similar to that of static indentation with a sharp indenter, the material behavior can be readily analyzed by employing indentation fracture mechanics [24–26]. Erosion of brittle materials by hard, sharp particles is generally thought to result from elastic/plastic fracture. Under low pressure and small erosion angle, the impacting effect of AWJ is weakened and the shearing effect is enhanced. In the erosion process of a single abrasive particle, the mechanics behavior of the workpiece material caused by the erosion effect is similar as that caused by the scratching of a moving sharp indenter or a single-point diamond tool. To analyze the details of the stress field outside the inelastic zone, the Boussinesq and Cerruti solutions can be used for concentrated point forces rather than the actual contact pressure distribution action between the indenter and the specimen surface. A coordinate system is established in Fig. 1.

ratio of Ft to Fn, dimensionless

bfij(y,f)+lgij(y,f)cu angular variation of the stress components or the principal stresses for a Poisson’s ratio u, dimensionless

Combining Boussinesq solution and Cerruti solution, the stress field for erosion process by a single abrasive particle can be obtained [27]:   8 Fn 1  2n 3 > >  sin2 f cos f s rr ¼ > 2 > > pR 2ð1"þ cos fÞ 2 > # > > > > F t cos y ð1  2nÞ sin f 3 > 3 >  sin f þ > > > 2 ð1 þ cos fÞ2 2 pR2 > > >   > > > F n ð1  2nÞ 1 F t cos y > > cos s ¼ f  þ > yy > 2 1 þ cos f > p R2 pR2 > > " !# > > > ð1  2nÞ 1 > < sin f 1   2 ð1 þ cos fÞ2 >     > > > Fn 3 F t cos y 3 > 3 2 > cos sin s ¼  f f cos f  > zz > 2 > pR2 2 pR2 > >    > > F 3 F t cos y 3 > n 2 2 > > cos sin s  f sin f f cos f  rz ¼ > > 2 > pR2 2" pR#2 > > > > > F t sin y ð1  2nÞ sin f > > > > sry ¼ > 2 ð1 þ cos fÞ2 pR2 > > > : szy ¼ 0

(1)

where Fn is the vertical load, Ft is the horizontal load, u is the poisson’s ratio of material, R, y and F represent the polar coordinates and srr, syy, szz, srz, sry and szy are respectively, stress components. The ratio of Ft to Fn, l, can be defines by



Ft ¼ cot a Fn

(2)

Fn

X

Ft θ r Φ R Y

σrr Z

σθθ

σrz σzz

Fig. 1. Coordinate system for erosion process by a single abrasive particle.

ARTICLE IN PRESS H.T. Zhu et al. / International Journal of Machine Tools & Manufacture 49 (2009) 569–578

where a is the erosion angle. So, Eq. (1) can be expressed as follows:

0.4

0 -0.2 -0.4 -0.6 -0.8

σ33/(Fn/(πR2))

-1.2 -1.4 -1.6 -100

-80

-60

-40

-20

0

20

40

60

80

100

Erosion angle (deg) Fig. 2. Angular variations of principal normal stresses (erosion angle of 901, Poisson’s ratio of 0.28).

1.2 1

(4)

σ13/(Fn/(πR2))

0.8 fij(θ,φ)+λgij(θ,φ)

And the expression for the principal shear stresses is [28]: 8 s ¼ 1ðs  s33 Þ > > < 13 2 11 s12 ¼ 12ðs11  s22 Þ > > : s23 ¼ 1ðs22  s33 Þ

σ22/(Fn/(πR2))

-1

(3)

According to the above equation, the expression for the principal normal stresses is given by rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 8 s  s 2 srr þ szz > rr zz > > þ s ¼ þ s2rz > 11 > 2 2 < s22 ¼ syy rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi > s  s 2 > > srr þ szz rr zz > >  þ s2rz : s33 ¼ 2 2

σ11/(Fn/(πR2))

0.2

fij(θ,φ)+λgij(θ,φ)

  8 Fn 1  2n 3 2 > > sin  s ¼ f cos f rr > > þ cos fÞ 2 > pR2 2ð1 > > " #) > > > ð1  2nÞ sin f 3 > 3 > l cos y  f þ sin > > > 2 ð1 þ cos fÞ2 2 > > >    > > > F n ð1  2nÞ 1 > > cos s ¼ f  > yy > 1 þ cos f > pR2 " 2 > > !#) > > > ð1  2 n Þ 1 > < sin f 1  þl cos y 2 ð1 þ cos fÞ2 >     > > > 3 3 > s ¼ Fn > sin f cos2 f  cos3 f  l cos y > zz 2 > 2 > pR  2 > >    > > Fn 3 3 > 2 2 > > cos sin s ¼  f sin f l cos y f cos f  rz > > 2 > pR2 " 2 > # > > > > F ð1  2nÞ sin f > > s ¼ n l sin y > > > ry pR2 2 ð1 þ cos fÞ2 > > > : szy ¼ 0

571

0.6 0.4 2

σ12 /(Fn/(πR ))

0.2 (5)

0

2

2

σ23/(Fn/(πR )) For an isotropic material of Poisson’s ratio u, the stress components and the principal stresses can be rewritten in general form as follows:

sij ¼

Fn

pR2

½f ij ðy; fÞ þ lg ij ðy; fÞu

(6)

where bfij(y,f)+lgij(y,f)cu is the angular variation of the stress components or the principal stresses for Poisson’s ratio u. During erosion by an abrasive particle, it can be found from Eq. (6) that an elastic stress field is generated and its magnitude decays with the distance R from the centre of the point load. Above the stress field is an elastic solution for the point load. So, it should be noted that the stress field contains a singularity at the contact origin R ¼ 0, where the sharp indenter induces a zone of plastic deformation underneath the contact point of the indenter [28–30]. Yoffe developed a model to describe the elastic/plastic stress fields and fractures generated under a quasi-static indentation by a pointed indenter in brittle materials [31]. This model treated the indentation plastic zone as a surface inclusion in a half-space and considered the indentation stress field in the fully loaded state to be the summation of the plastic stress field and the elastic stress field. After unloading, the residual stress field, which drives the lateral cracks, is formed by the plastic stress field. Several investigations have demonstrated that Yoffe’s model can explain the experimental phenomenon of indentation-induced cracking [32] and agree with the FEM results [33].

-0.2 -100 -80

-60

-40

-20 0 20 40 Erosion angle (deg)

60

80

100

Fig. 3. Angular variations of principal normal stresses (erosion angle of 901, Poisson’s ratio of 0.28).

The erosion load and the erosion angle are the key factors affecting the stress field. If the erosion angle is constant, the magnitudes of the erosion stresses beneath erosion zone are decided by the erosion kinetic energy of an abrasive particle, which decides the magnitudes of the erosion loads, Fn and Ft. When the erosion kinetic energy of an abrasive particle is constant, stress field distribution and the magnitudes of stress components can be affected by l, which can be controlled by the erosion angle a. When the erosion angle is 901 and the horizontal load Ft ¼ 0, the stress field can be obtained from the classical Boussinesq solution. The angular variation of the principle normal stresses and the principle shear stresses are plotted in Figs. 2 and 3. The distribution of the stresses is axially symmetrical. The principal normal stress of s11 is tensile and the maximum values appear in F ¼ 7901. The principal normal stress of s33 is compressive and the maximum values appear in F ¼ 01. The principal normal stress of s22 is tensile within a conical region Fo51.81 below the point loading and is compressive outside this region. The maximum values of the principal shear stresses of s13 and s23 appear in F ¼ 01, while the maximum value of s12 appears in

ARTICLE IN PRESS 572

H.T. Zhu et al. / International Journal of Machine Tools & Manufacture 49 (2009) 569–578

1.5

1.2 2

σ13/(Fn/(πR ))

1

1

σ11 /(Fn /(πR2))

0.8 fij(θ,φ)+λgij(θ,φ)

fij(θ,φ)+λgij(θ,φ)

0.5 0 -0.5

σ22 /(Fn /(πR2))

-1

σ33 /(Fn /(πR2))

0.6 0.4 2

σ12/(Fn/(πR ))

0.2

-1.5

0

-2 -2.5 -100

2

σ23/(Fn/(πR ))

-80

-60

-40

-20

0

20

40

60

80

100

Erosion angle (deg) Fig. 4. Angular variations of principal normal stresses (erosion angle of 451, Poisson’s ratio of 0.28).

-0.2 -100 -80

-60

-40

-20 0 20 40 Erosion angle (deg)

60

80

100

Fig. 5. Angular variations of principal normal stresses (erosion angle of 451, Poisson’s ratio of 0.28).

F ¼ 7901. Comparing magnitudes of the angular variation of the principle stresses, it is concluded that cracks induced by tensile stress are mainly governed by s11 and shear flow and shear cracks of material induced by shear stress are mainly governed by s13. Observations of microcracking due to quasi-static normal point indentation under various experimental conditions have been previously made by many researchers [28–30,34]. Lawn and Swain summarized a general pattern [28]. First, in initial loading stage, the sharp indenter induces a zone of plastic deformation about the contact point. And then, at a certain critical indenter load, a crack suddenly starts of below the contact point. These cracks are commonly termed median/radial cracks. During the unloading of the indenter, the residual stress generated as a result of incompatibility between the plastic deformation zone and the surrounding elastic material initiates cracks. These cracks are commonly termed as the lateral cracks. The lateral cracks extend mainly parallel to the material surface. The lateral cracks are assumed to be mainly responsible for material removal in erosion and abrasion, while the median/radial cracks are associated with strength degradation [35]. It is known that there exists a critical load over which the cracks begin to form. In the case of small erosion angle, the erosion is not normal but inclined to the surface of workpiece. So, the stress field distribution deflects from the axial symmetrical position to the direction of the horizontal load because of the effect of the horizontal load Ft, as shown in Figs. 4 and 5. The principal normal stress of s11 has the maximum value in F ¼ 901, while the principal normal stress of s33 has the maximum value about F ¼ 401. The maximum values of the principal shear stresses of s13 and s23 appear about F ¼ 401. The tensile stress behind the moving abrasive particle can engender cracks in the material near surface. These cracks are inclined to propagating in the plane of F ¼ 901. In front of the moving abrasive particle, the material is subjected to a combination action of shear and compressive stresses. The combination action creates yielding of the material under the indenter and the yielding process continues until the yielded zone attains its stable preferred state [31]. The deformations of material include shear and compressive deformation in most ceramics [36]. According to the above analysis of erosion stress, it can be concluded that the vertical component of the erosion load deals with crack nucleation and propagation, while the horizontal component of the erosion load facilitates microcutting in front of moving abrasive particle.

Fig. 6. Erosion by a single particle of AWJ on small erosion angle.

Based on the Yoffe’s model, several investigations on stress distribution of a moving indentation have been conducted [36,37]. Nevertheless, for a moving indentation, there is as yet no model that can completely describe the elastic/plastic stress fields under the indenter and further investigation is needed. Experimental study for the erosion process of a single abrasive particle can reveal further the material deformation and removal mechanisms. In experimental study, it is found that for hard– brittle materials, there are two different modes of material removal by a single abrasive particle depending upon the erosion kinetic energy that the single abrasive particle exerts on the material surface [38]. If the erosion kinetic energy is larger than the threshold for crack initiation, material removal is mainly caused by the microcrack propagation. However, when the erosion kinetic energy is less than the threshold, the plastic flow is the dominant material removal mechanism in ductile material. When a single abrasive particle erodes the surface of workpiece, erosion streak can be formed. Being analogous with moving indentation or scratch experiment, the residual imprinting and the extension of cracks can be formed during erosion process by single abrasive particle. The erosion process of a single abrasive particle on workpiece surface at a small erosion angle is shown in Fig. 6. When the abrasive particle just touches the material surface, the erosion is in elastic scratch stage. In the elastic scratch stage, elastic deformation occurs between the abrasive and the workpiece surface but there is no material removal. With an increase in the depth pressed into workpiece surface by the

ARTICLE IN PRESS H.T. Zhu et al. / International Journal of Machine Tools & Manufacture 49 (2009) 569–578

573

Fig. 7. Erosion surface with erosion angle of 601.

Table 1 Workpiece materials. Materials

Elastic modulus E (GPa)

Hardness Hv (GPa)

Fracture toughness KIC (Mpa m1/2)

Workpiece dimension (mm)

Surface roughness Ra (mm)

Silicate glass 96% Al2O3 Si3N4

69 300 311

5.6 18 16

0.8 4 8

10  20 10  20 15  20

0.334 0.393 0.167

abrasive particle, the plastic distortion appears under the erosion site. In this stage, there exists a small pile-up of material on two sides of the erosion streak and in front of the abrasive particle and microcutting may occur. This is the plowing and cutting stage in which micromaterial removal can be achieved by ductile flow and the erosion mechanism is ductile erosion. When the depth of the erosion streak exceeds a critical value, the fracture can appear in material and material removal mechanism is changed from the initial ductile flow to the brittle fracture. In this stage, the erosion mechanism is fracture erosion. Different from moving indentation or scratch experiment, when the vertical component of the erosion kinetic energy is consumed completely, the abrasive particle will rebound off the workpiece surface. The depth of the erosion streak depends on the vertical component of the erosion kinetic energy but not the general erosion kinetic of the abrasive particle. By controlling the vertical component of the erosion kinetic energy less than a critical value, ductile erosion without crack formation can be achieved. The horizontal component of the erosion kinetic energy will play an important role in promoting microcutting during the erosion process. Fig. 7 shows the erosion surfaces, which are eroded using a jet pressure of 120 MPa, an erosion angle of 601 and alumina abrasive with a grain diameter of 38 mm. Despite considerable brittle fractures, ductile erosion streaks can still be observed in erosion surface of silicate glass while plastic deformation is still observed in alumina ceramics. (The fracture surfaces are light and white with obvious cracks in SEM photograph. On the contrary, the ductile erosion or plastic surfaces are always dark in SEM photograph, as indicated by arrows in Fig. 7.) It indicates that the ductile erosion mechanism does exist in a small angle erosion. It is possible that an erosion machining method dominantly characterized by the ductile erosion can be achieved using smaller erosion angle and lower jet pressure. Ductile erosion can achieve micromaterial removal and the eroded surface is smooth without any fracture. By using ductile erosion method with low pressure and small erosion angle, the precision surface machining can be carried out by AWJ.

3. Experimental procedure In order to analyze the feasibility of precision surface machining by AWJ, a group of polishing experiments is performed. Three kinds of hard–brittle material were chosen, namely, silicate glass, 96% alumina and silicon nitride. The workpiece material properties are shown in Table 1. Before polishing by AWJ, the workpieces were lapped by using abrasive B4C with a size of 10 mm. The same B4C abrasive was also used in the present polishing experiment. The material properties of B4C are shown in Table 2. A precision abrasive waterjet micromachining system is employed. This machining system can generate a jet with a pressure of 2–15 MPa and a diameter of 0.3 mm. The precision abrasive waterjet micromachining system has a three-axis motion system, which numerically controls the motion of the nozzle. The polishing conditions are listed in Table 3. The threshold kinetic energies of ductile erosion for silicate glass, 96% alumina and silicon nitride are 4.7813  108, 3.4255  106 and 4 3.3806  10 J, respectively [39]. Under the erosion conditions of this study, the vertical component of the average erosion kinetic energy obtained by B4C particles is 2.5135  109 J. It indicates that the machining mechanism is the ductile erosion in these polishing experiments. Polishing surfaces of workpieces were observed with a JSM6700F scanning electron microscope (SEM). A NanoScope III, a type of atomic force microscopy (AFM), was used to measure the surface profile of workpiece and the results were analyzed by a NanoScopr (R) III Version 5.30r3.sr3 analysis software.

4. Results and discussion Comparing the polishing and lapping surfaces of the three materials, as shown in Figs. 8–10, it is found that the surfaces of workpieces are improved greatly by AWJ polishing. The lapping surfaces of the three kinds of materials are coarse and have some fracture. In the lapping surface of silicate glass as shown in

ARTICLE IN PRESS 574

H.T. Zhu et al. / International Journal of Machine Tools & Manufacture 49 (2009) 569–578

Table 2 Abrasive. Abrasive

Content (%)

Hardness Hv (GPa)

Density r (g/cm3)

Particle size dp (mm)

B4C

90–92

54–63

2.52

10

Table 3 Experimental conditions. Waterjet pressure (MPa)

Standoff distance (mm)

Erosion angle (deg.)

Flow rate (mg/min)

Traverse rate (mm/s)

Lateral spacing (mm/pass)

15

2

15

120

0.5

0.5

 3 times polishing for every workpiece

Fig. 8. Silicate glass machined surface. (a) Lapped and (b) polished.

Fig. 9. 96% Al2O3 ceramic machined surface. (a) Lapped and (b) polished.

Fig. 8(a), numerous brittle fractures with considerable overlapping of damage zones and residual cracks can be found. Fig. 9(a) shows lapping surface of alumina in which there are considerable fracture damages characterized by intergranular cracks and exposed pores. Because of high strength, high toughness and high compactness, no residual cracking and pore are evident in the lapping surface of silicon nitride as shown in

Fig. 10(a). However, it is readily observed that some pits are formed by brittle fracture. After AWJ polishing, the surfaces of all kinds of workpieces become fine and there is almost no fracture in polishing surfaces which are mainly composed of ductile flow surface. And the polishing surfaces have very nice surface integrity. On the polished surfaces of silicate glass and silicon nitride, some ductile

ARTICLE IN PRESS H.T. Zhu et al. / International Journal of Machine Tools & Manufacture 49 (2009) 569–578

Fig. 10. Si3N4 ceramic machined surface. (a) Lapped and (b) polished.

Fig. 11. AFM analysis of silicate glass polished surface by AWJ. (a) 3D profile and (b) surface rovghness.

575

ARTICLE IN PRESS 576

H.T. Zhu et al. / International Journal of Machine Tools & Manufacture 49 (2009) 569–578

Fig. 12. AFM analysis of 96% Al2O3 ceramic polished surface by AWJ. (a) 3D Profile and (b) surface rovghness.

erosion streaks can be found as shown in Fig. 8(b) and Fig. 10(b) and the depth of streaks of the three materials is very small, while the plastic deformation is evident on the surface of 96% alumina as shown in Fig. 9(b). These indicate that the materials are removed by microductile flow. The characteristics of the ductile erosion are highly advantageous to precision surface machining for hard–brittle materials. A group of AFM analyses including 3-D profile and surface roughness on AWJ polishing surface was performed. The surface roughness of silicate glass achieves the Ra ¼ 93.195 nm in the measured area and the averaged Ra ¼ 40–46 nm on the two perpendicular sections. The surface roughness of aluminum achieves the Ra ¼ 131.22 nm in measured area and averaged Ra ¼ 40–50 nm on the two perpendicular sections. The surface roughness of silicon nitride achieves Ra ¼ 38.616 nm in measured

area and averaged value Ra ¼ 26–38 nm on the two perpendicular sections. The polished surfaces of the three materials have a good finish. The erosion streaks of silicate glass are big and deep because of low fracture toughness and strength. So, the surface roughness of silicate glass is also higher as shown in Fig. 11. Comparing the three materials, the surface roughness of aluminum is the highest as shown in Fig. 12. The primary cause is that considerable material flaws, such as pores and intergranular cracks, exist in aluminum. The surface roughness of silicon nitride is the least one. Because the silicon nitride has high fracture toughness and high compactness, the erosion streaks are small and shallow. The material is removed by microductile flow, so the erosion polished surface has a very good finish. The AFM analysis of silicon nitride is shown in Fig. 13.

ARTICLE IN PRESS H.T. Zhu et al. / International Journal of Machine Tools & Manufacture 49 (2009) 569–578

577

Fig. 13. AFM analysis of Si3N4 ceramic polished surface by AWJ. (a) 3D Profile and (b) surface rovghness.

According to the above analyses, it can be concluded that the surfaces of workpieces are improved greatly by AWJ polishing but the achieved surface quality level is affected by some other factors such as material property, machining flaw and so on. The materials having high fracture toughness, high strength, less material flaws and machining flaws are prone to achieving a low surface roughness.

5. Conclusions In this paper, a stress analysis of erosion process by angular abrasive particle has been undertaken. It can be concluded that the vertical component of the erosion load results in crack nucleation and propagation while the horizontal component of the erosion load facilitates microcutting in front of moving

abrasive particle. For hard–brittle materials, there are two different modes of material removal by a single abrasive particle depending upon the erosion kinetic energy that the single abrasive particle exerts on the material surface, and these are fracture erosion and ductile erosion. By controlling the vertical component of the erosion kinetic energy less than a critical value, the ductile erosion without crack formation will be achieved. The horizontal component of the erosion kinetic energy will play an important role in promoting microcutting during the erosion process. Ductile erosion can achieve micromaterial removal and the eroded surface is smooth without any fracture. The characteristics of the ductile erosion are highly advantageous to precision surface machining for hard–brittle materials. By using ductile erosion method with low pressure and small erosion angle, the precision surface machining can be carried out by AWJ.

ARTICLE IN PRESS 578

H.T. Zhu et al. / International Journal of Machine Tools & Manufacture 49 (2009) 569–578

In this study, a group of polishing experiment for brittle material by AWJ was conducted and the feasibility of polishing by AWJ was investigated. Polishing experiment results indicate that AWJ has a great potential to be used as a precision surface machining technology. Moreover, according to SEM and AFM analysis, it can be concluded that the surfaces of workpieces are improved greatly by AWJ polishing but the achieved surface quality level is affected by some other factors e.g. as material property and machining flaw. The materials having high fracture toughness, high strength, less material flaws and machining flaws are prone to yield a low surface roughness.

Acknowledgement This project is supported by Outstanding Young Scholar Science Foundation of NSFC—International Collaborative Research Foundation (no. 50328505) and Key Project of Shandong Natural Science Foundation (Z2005F02). References [1] C.Z. Huang, R.G. Hou, J. Wang, Y.X. Feng, The effect of high pressure abrasive water jet cutting parameters on cutting performance of granite, Key Engineering Materials 304–305 (2006) 560–564. [2] J. Wang, Abrasive Waterjet Machining of Engineering Materials, Trans Tech Publications, Switzerland, 2003. [3] C.Z. Huang, J. Wang, Y.X. Feng, H.T. Zhu, Recent development of abrasive water jet machining technology, Key Engineering Materials 315–316 (2006) 396–400. [4] S. Xu, J. Wang, A study of abrasive waterjet cutting of alumina ceramics with controlled nozzle oscillation, International Journal of Advanced Manufacturing Technology 27 (7–8) (2006) 693–702. [5] M. Hashish, Cutting with abrasive waterjets, Mechanical Engineering 106 (3) (1984) 60–69. [6] J. Zeng, T.J. Kim, Erosion model of polycrystalline ceramics in abrasive waterjet cutting, Wear 193 (2) (1996) 207–217. [7] H. Liu, J. Wang, N. Kelson, R.J. Brown, A study of abrasive waterjet characteristics by CFD simulation, Journal of Materials Processing Technology 153–154 (1–3) (2004) 488–493. [8] M. Hashish, Optimization factors in abrasive-waterjet machining, Journal of Engineering for Industry, Transactions of the ASME 113 (1) (1991) 29–37. [9] R. Kovacevic, M. Fang, Modeling of the influence of the abrasive waterjet cutting parameters on the depth of cut based on fuzzy rules, International Journal of Machine Tools & Manufacture 34 (1) (1994) 55–72. [10] L. Chen, E. Siores, W.C.K. Wong, Kerf characteristics in abrasive waterjet cutting of ceramic materials, International Journal of Machine Tools & Manufacture 36 (11) (1996) 1201–1206. [11] J. Wang, Predictive depth of jet penetration models for abrasive waterjet cutting of alumina ceramics, International Journal of Mechanical Sciences 49 (3) (2007) 306–316. [12] J. Wang, T. Kuriyagawa, C.Z. Huang, An experimental study to enhance the cutting performance in abrasive waterjet machining, Machining Science and Technology 7 (2) (2003) 191–207. [13] J. Wang, D.M. Guo, The cutting performance in multipass abrasive waterjet machining of industrial ceramics, Journal of Materials Processing Technology 133 (3) (2003) 371–377. [14] E. Siores, W.C.K. Wong, L. Chen, J.G. Wager, Enhancing abrasive waterjet cutting of ceramics by head oscillation techniques, CIRP Annals—Manufacturing Technology 45 (1) (1996) 327–330.

[15] M. Hashish, Investigation of milling with abrasive-waterjets, Journal of Engineering for Industry, Transactions of the ASME 111 (2) (1989) 158–166. [16] J. Zeng, T.J. Kim, Erosion model for abrasive waterjet milling of polycrystalline ceramics, Wear 199 (2) (1996) 275–282. [17] Y.X. Feng, C.Z. Huang, J. Wang, R.G. Hou, X.Y. Lu, An experimental study on milling Al2O3 ceramics with abrasive waterjet, Key Engineering Materials 339 (2007) 500–504. [18] Y.X. Feng, C.Z. Huang, J. Wang, X.Y. Lu, H.T. Zhu, Surface characteristics of ceramics milled with abrasive waterjet technology, Key Engineering Materials 329 (2007) 335–340. [19] Z. Guo, M. Ramulu, Investigation of displacement fields in an abrasive waterjet drilling process: Part 1. Experimental measurements, Experimental Mechanics 41 (4) (2001) 375–387. [20] Z. Guo, M. Ramulu, Investigation of displacement fields in an abrasive waterjet drilling process: Part 2. Numerical analysis, Experimental Mechanics 41 (4) (2001) 388–402. [21] M. Hashish, Precision drilling of composites with abrasive-waterjets, American Society of Mechanical Engineers, Materials Division (Publication) MD 45, Machining of Advanced Composites (1993) 217–225. [22] M. Hashish, J. Whalen, Precision drilling of ceramic-coated components with abrasive-waterjets, Journal of Engineering for Gas Turbines and Power, Transactions of the ASME 115 (1) (1993) 148–154. [23] H.C. Meng, K.C. Ludema, Wear models and predictive equations: their form and content, Wear 181–183 (2) (1995) 443–457. [24] J.E. Ritter, P. Strzepa, K. Jakus, L. Rosenfeld, K.J. Buckman, Erosion damage in glass and alumina, Journal of the American Ceramic Society 67 (11) (1984) 769–774. [25] M. Buijs, Erosion of glass as modeled by indentation theory, Journal of the American Ceramic Society 77 (6) (1994) 1676–1678. [26] P.J. Slikkerveer, P.C.P. Bouten, F.H. in’t Veld, H. Scholten, Erosion and damage by sharp particles, Wear 217 (2) (1998) 237–250. [27] B. Lin, Theoretical and Experimental Studies on the Ultra-Precision Grinding of Engineering Ceramics, Tianjin University, Tianjin, China, 1999 Ph. D. Dissertation. [28] B.R. Lawn, M.V. Swain, Microfracture beneath point indentations in brittle solids, Journal of Materials Science 10 (1) (1975) 113–122. [29] B.R. Lawn, A.G. Evans, D.B. Marshall, Elastic/plastic indentation damage in ceramics: the median/radial crack system, Journal of the American Ceramic Society 63 (9–10) (1980) 574–581. [30] D.B. Marshall, B.R. Lawn, A.G. Evans, Elastic/plastic indentation damage in ceramics: the lateral crack system, Jourmal of the American Ceramic Society 65 (11) (1982) 561–566. [31] E.H. Yoffe, Elastic stress fields caused by indenting brittle materials, Philosophical Magazine A: Physics of Condensed Matter, Defects and Mechanical Properties 46 (4) (1982) 617–628. [32] R.F. Cook, G.M. Pharr, Direct observation and analysis of indentation cracking in glasses and ceramics, Journal of the American Ceramic Society 73 (4) (1990) 787–817. [33] G. Feng, S. Qu, Y. Huang, W.D. Nix, An analytical expression for the stress field around an elastoplastic indentation/contact, Acta Materialia 55 (9) (2007) 2929–2938. [34] B.R. Lawn, A.G. Evans, Model for crack initiation in elastic/plastic indentation fields, Journal of Materials Science 12 (11) (1977) 2195–2199. [35] S. Malkin, J.E. Ritter, Grinding mechanisms and strength degradation for ceramics, Journal of Engineering for Industry, Transactions of the ASME 111 (2) (1989) 167–174. [36] Y. Ahn, T.N. Farris, S. Chandrasekar, Sliding microindentation fracture of brittle materials: role of elastic stress fields, Mechanics of Materials 29 (3-4) (1998) 143–152. [37] H.Z. Wu, S.G. Roberts, B. Derby, Residual stress distributions around indentations and scratches in polycrystalline Al2O3 and Al2O3/SiC nanocomposites measured using fluorescence probes, Acta Materialia 56 (1) (2008) 140–149. [38] H. Wensink, M.C. Elwenspoek, A closer look at the ductile–brittle transition in solid particle erosion, Wear 253 (9–10) (2002) 1035–1043. [39] H.T. Zhu, Study on Erosion Mechanisms and Polishing Technology of HardBrittle Materials Machined with Precision Abrasive Waterjet, Shandong University, Jinan, P.R. China, 2007 PhD Dissertation.