Open Access
Add to BibSonomyAbstract
The subsurface damages (SSD) of fused silica developed during deterministic small tool polishing are experimentally investigated in this study. A leather pad (i.e., poromeric) is validated to be nearly SSD-free and superior to pitch and polyurethane. Rough abrasives are found to obviously increase SSD depth, and a leather pad can efficiently suppress the adverse effect of rough abrasives. The SSD depth induced by pitch and polyurethane pads (with rough abrasive) ranges from 0.77 to 1.49μm (~1/7-1/5 of abrasive size). High pressure, low velocity and slurry concentration can slightly increase SSD depth. Material removal rate of leather pad is also validated to be comparable with polyurethane and much higher than pitch tool; surface roughness polished by leather pad is Ra = 1.13nm, which is close to that of pitch but much better than polyurethane.
© 2014 Optical Society of America
1. Introduction
Computer controlled subaperture polishing technique is widely used in manufacturing precision optical segments; in particular, the small tool polishing technique developed from 1970s [1–4]. Compared with other advance polishing techniques [5–9], small tool polishing technique provides high removal rates and various tool sizes, as well as an excellent smoothing effect to mid-spatial frequency errors. Although its edge effect still exists and removal stability is relatively low, small tool polishing is still widely used in pre- and fine-polishing of optical segments, as well as in combination with magnetorheological finishing (MRF) [10] and ‘Precessions’ bonnet polishing [11].
Subsurface damages (SSDs), which include pits, cracks and scratches, are rarely considered in small tool polishing. SSD limits applications of small tool polishing in some cases regardless of whether this technique is applied at the end of the manufacturing process or not. For intense laser systems, SSD degrades laser-induced damage threshold, increases mechanical weakness, and then induces macroscopic damage [12,13]. For image and telescope systems, SSD diminishes longtime stability, coating quality and image capability [14]. In ultra-precision cases (e.g., lithography lens), optics polished by small tool should be further figured by IBF to remove SSD and improve surface accuracy to nanometer level, which requires a long time because of low material removal rate. Therefore, the SSD features of small tool polishing should be investigated and improved to extend its applications in some critical cases.
SSD in grinding and lapping has been extensively investigated in the past decades [15–32]. SSD is formed by mechanical interaction of tool, abrasive and optical segments (two-body abrasion for fixed abrasive and three-body abrasion for loose abrasive [15]). Thus, SSD is closely related to the mechanical properties of optical segments (e.g., elastic modulus, hardness and fracture toughness etc.), abrasive type (hardness) and shape (dull or sharp). Lambropoulos et al. considered that both median and lateral cracks can form when the load on an abrasive particle exceeds a critical value [16,17]. The lateral crack removes material and the median crack induces features similar to SSD. Based on indentation of sharp indenter and microindentation mechanics [17], they also derived theoretical equations of median and lateral crack depths, which are correlative to the mechanical properties of optical segments.
SSD depth also highly relates with manufacturing parameters. It is found to be direct proportional to abrasive size and load on per abrasive. Lambropoulos el al [18]. collected data on optical glasses during microgrinding and lapping, and they found that SSD depth is bounded according to Eq. (1), where,denotes the abrasive size in micrometers. Tonnellier et al. [21] investigated the grinding mode and used wedge polishing technique to assess the amount of damage under a ground ULE and Zerodur with different brittleness; and they also found the number of defects apparent at different depths beneath the surface is a combined function of ‘process’ related and ‘machine dynamics’ related damage. Suratwala and Miller et al. [22–24] investigated and measured the distribution and characteristics of SSD formed during grinding and polishing. They found that only rough abrasives are being mechanically loaded; and the SSD depth increases with the load on per abrasive. The adverse effect of rough abrasives is also studied in detail [23]. Slurry chemistry is also a significant factor [25,26]. Neauport et al. [27,28] investigated loose abrasive lapping and diamond grinding of fused silica by HF etching. They found that the relation between SSD depth and abrasive size for loose abrasive lapping is SSD = 0.74L1.04. More recently, J. Neauport et al. [29], Laheurte et al. [30] and Blaineau et al. [31] developed new non-destructive methods for SSD measurement, such as confocal fluorescence microscopy and Abbott–Firestone curves. Johnson and Kim et al. [32] reported a 1.1 × ratio between SSD depth and abrasive size for 3M Trizact bound-abrasive lapping, which is superior to their results on loose abrasive lapping (~1.4 × ratio).
SSD depth is also closely related to surface roughness (SR). Experimental ratio of SSD depth and Rt (peak to valley of SR) for diamond wheel grinding, micro-grinding, loose or bound abrasive lapping have been proposed in several studies [16,19,24,25,27,28] (see Table 1), which can estimate SSD depth in a non-destructive manner. The difference in these ratios is mainly caused by the difference in manufacturing methods, properties of material and measurement methods for SR and SSD depth.
Table 1. Previous Researches on SSD Depth and SR
Polishing can be considered as an extension of lapping and micro-lapping, in which residual SSD of previous grinding and lapping should be removed, and leaves minimal newly produced SSD. However, the material removal mechanism of polishing is not mechanically dominated as grinding and lapping, but is a mixture of mechanical and chemical removal as well as surface flow [33]. Therefore, the SSD character of polishing can be essentially different. However, except for the work of Suratwala et al. [23] on SSD induced by rotary polishing machine with pitch or polyurethane pads, there are few researches have focused on SSD developed in polishing, especially the deterministic small tool polishing.
In this study, SSD of fused silica developed during deterministic small tool polishing with three pads (pitch, polyurethane and leather), various abrasive slurries (cerium oxide: CeO2), pressures and velocities are investigated and optimized to achieve minimal SSD depth, respectively. The SSD depth is measured by virtue of the combination of magnetorheological ðnishing (MRF) taper [19,20,22,23], hydrofluoric acid (HF) corrosion, profile measurement and micro-examination. In particular, leather is proven to be an excellent pad, which is nearly SSD-free and has considerable removal rate and fine surface roughness.
2. Experimental procedures
2.1 Sample preparation
Round fused silica (JGS2: ultraviolet grade fused silica used in spectral band of 220nm-2500nm, produced by Shanghai Xinhu Glass Ltd.) with diameter of 150mm and thickness of 10mm are used. Fused silica is a widely used optical material for various optical systems, especially for intense laser systems. It processes a 73.0MPa Young’s modulus, 0.17 Poisson ratio, 6.6GPa hardness, 1.0MPam1/2 fracture roughness, 2200kg/m3 density and 0.56E−6/K thermal expansion coefficient. The samples are primarily fully polished by double-axis polishing machine with a Φ200mm pitch pad and CeO2 slurry (slurry 2# described in next subsection) to remove damages produced during previous grinding and lapping. This sample preparation would generate SSD with depth ~1μm (see subsection 3.4), which can be readily removed by subsequent small tool polishing process, so that not disturb the measurement results of SSD depth.
2.2 Small tool polishing process
Small tool polishing processes are conducted on a small tool polishing machine, namely, JR-1800 (see Fig. 1), with a polishing volume of 1840mm × 2096mm × 603mm [34]. The polishing tool adopts planetary structure, by which the spinning axis can orbit around an orbital axis, producing a Gaussian-shape tool influence function (TIF) [34]. The polishing pad has an aperture from 20mm to 200mm, which produces a relatively larger removal area compared with other polishing techniques. The samples are uniformly polished to remove ~5μm (much deeper than the per-existing SSD after sample preparation) in the central portion (see Fig. 2) by JR-1800 with predefined conditions, thereby ensuring that previous SSD is completely removed and only leaving SSD caused by the small tool polishing. It should be noted that even the per-existing SSD after sample preparation are not completely removed by small tool polishing, it is believed that it is weakened by small tool polishing, thus, the residual SSD is much shallower than the new SSD developed by small tools, so that it would not influence the measurement of SSD depth.
Three pads, namely polishing pitch, polyurethane and polishing leather, are used in this study. Polishing pitch [Fig. 3(a)] processes a relatively rigid but smooth interface compared with polyurethane and leather. It acts as a highly viscous Newtonian ñuid at long time scales. When it undergoes shear motion that is proportional to the shear stress, it flows to conform to the shape of optical segments. This fluidity at long time scales ensures smoothness of the entire lap surface. Abrasives can be embedded into the pitch then plough surfaces to remove materials. Pitch pad is a mostly-used tool for precision polishing [35]. The pitch used in this study was supplied by Satisloh GmbH, with softening point ~70°C; needle penetration is ~0.01mm at 50g load, referenced at 20°C.
Polyurethane [Fig. 3(b)] is also a frequently-used material for small tool polishing [36]. It is a foam type pad and has numerous independent bubbles with top surfaces that are broken open for retaining coolant and compounds. It has a range of hardness values, and is available pre-grooved. Polyurethane is durable and can polish much faster than pitch, but may need periodic dressing. The polyurethane used in this study has Shore A hardness of ~75-80.
Polishing leather is an artificial pad material (i.e., engineered poromeric) with napped surface and numerous capillary interstices [Fig. 3(c)]. Its surface has a thin layer of free-standing stalks that act like a brush to soften contact. It is especially useful for polishing soft materials and for quick removal of sleeks, contamination, and stains without affecting figure [37]. Moreover, it has strong capability of water absorption. In our test, 100g leather can absorb ~180-200g slurry after sufficient immersion. Compared with pitch and polyurethane, polishing leather has a larger contact area and a softer working interface, which drives more loaded abrasives per unit area on surfaces.
Leather and polyurethane exhibit poor viscoelasticity and fluidity, thus, they cannot efficiently conform to the aspherical surfaces if they are directly attached to a steel base. Silica rubber or pitch can be used as a bonding agent and interlayer to improve the deformability of leather and polyurethane pads. In this study, 2mm thickness silica rubber is adopted to bond polyurethane and leather on a steel base. The configurations of three pads are summarized in Table 2. The pads used in every experiment are new to avoid cross interaction of different abrasives.
Table 2. Configurations of Three Pads used in This Study
Three kinds of CeO2 are used for comparison as listed in Table 3. Their purities are measured using X-ray diffraction (Bruker, D8 ADVANCE) and analyzed by MDI JADE software. Their micrographs [Figs. 4(a)-4(c)] are measured by a scanning electron microscope (SEM, FEI, NOVA NANOSEM 430). CeO2 1# suffers from agglomeration phenomenon [see Fig. 4(a)], and it is hard to distinguish the largest particle size. The particles of CeO2 2# and 3# shown in Fig. 4(b) and 4(c) are more separated, and large particles with sharp edges can be found, which can indicate the surficial topography of abrasives and approximately confirm the largest particle size. Their size distribution curves [Fig. 4(d)] are determined by a laser particle size analyzer (Malvern, Mastersizer 2000). During measurements process, the particles are vibrated by ultrasonic to avoid the agglomeration and sediment phenomenon, which can supply reliable results on particle size distribution. CeO2 1# (used for slurry 1#) processes medium particle diameter D50 = 1.10μm, D90 = 1.3μm, and purity 99.9%. The SEM micrograph [Fig. 4(a)] and the narrow size distribution curve [Fig. 4(d), blue curve] of this powder show relatively uniform size distribution and few rough abrasives. CeO2 2# (used for slurry 2#, 4# and 5#) has D50 = 3.2μm, D90 = 5.4μm, and purity 82%, doped by praseodymium (~16%) and other rare earth. Its SEM micrograph [Fig. 4(b)] shows numerous large abrasives, agreeing with its relatively wide size distribution curve [Fig. 4(d), green curve]. CeO2 3# (used for slurry 3#) has D50 = 4.3μm, D90 = 7.8μm, purity 91%, doped by ~7.8% praseodymium and other rare earth. Its SEM micrograph [Fig. 4(c)] shows more large abrasives, and its size distribution curve is the widest [Fig. 4(d), red curve]. These three kinds of CeO2 compose the five slurries as listed in Table 3, and pH values of five slurries are measured by a pH meter (with resolution 0.01 and accuracy ± 0.03), which are slightly alkaline.
Table 3. Configurations of Five Slurries Used in This Study
Fig. 4 (a) SEM micrograph of CeO2 1#, 10000 × magnification, one graduation 10μm; (b) SEM micrograph of CeO2 2#, 2000 × magnification, one graduation 50μm; (c) SEM micrograph of CeO2 3#, 2500 × magnification, one graduation 50μm; (d) size distribution curves of three kinds of CeO2.
2.3 SSD inspection and measurement
Destructive measurement of SSD depth needs a SSD-free polishing tool to produce a spot or taper without newly produced SSD. MRF is widely accepted as a SSD-free tool because material removal is induced by the shearing force of MR fluid [19,20,22,23]. For measuring SSD depth, segments after small tool polishing are then polished by MRF-180 to reveal the damage layer, under conditions: 8wt% CeO2 1#, 35wt% icon powder, 45wt% water, 10wt% glycerin and 2wt% additive, spinning velocity 300rpm, revolution velocity 30rpm, plunge depth of MR fluid 1mm. The MRF spot is extended in one-dimensional scanning mode with a 10mm stroke, generating a ~13mm × 5mm groove.
SSD is mostly obscured by the hydrolysis layer developed during polishing. In this study, segments are etched by diluted HF (5vol%) for 10min to reveal cracks under the hydrolysis layer. Then, they are cleaned by an ultrasonic cleaner (KunShan KQ3200B) for 5min to remove impurities and contaminants. The surface form of MRF groove is then measured by a coordinate measuring machine (CMM), with a resolution of 10nm. The data are saved as ‘.xyz’ format of Zygo interferometers and analyzed by MetroPro.
The cracks exposed along the taper at various depths are then inspected under an optical microscope with a 60 × objective, relating to a field of view (FOV) 55 × 68μm. Surface morphologies are recorded using a 720 × 576 CCD. With the help of the two-dimensional translation stage and screw micrometer on the microscope, the objective can be aligned to any position of the taper and record the position where microcracks disappear and emerge.
A surface form of MRF groove is presented in Fig. 5 for illustrating the determination of SSD depth. The groove is inspected under a microscope and the surface morphologies along the four lines [Fig. 5(a)] are recorded using a CCD camera. The profiles of four lines are shown in Fig. 5(b). Figure 6 illustrates surface morphologies along line AA’ as removal depth increasing for the fused silica polished by a polyurethane pad, with slurry 2#, 0.10MPa pressure and 300rpm tool spinning velocity. The cracks degrade gradually and the last crack [marked by the ellipse in Fig. 6(f)] is found at a depth of 0.79μm. Surface morphologies that deeper than 0.79μm are found to be SSD-free as shown in Figs. 6(g) and 6(h). Noteworthy, the CCD has several defective pixels that appear at all surface morphologies [e.g., marked by dotted circles in Figs. 6(g) and 6(h)], which should be neglected throughout this study.
Fig. 5 (a) Surface form of the MRF groove; (b) four profiles used for inspecting SSD; (c) profile of line AA’ and two inspectors; (d) the position difference of two inspectors in X and Y direction.
Fig. 6 Surface morphologies as the removal depth increasing (a) 0.0μm; (b) 0.16μm; (c) 0.32μm; (d) 0.48μm; (e) 0.63μm; (f) 0.79μm; (g) 0.95μm; (h) 1.11μm.
Accurate measurement of SSD depth is determined by the alignment error of the segment on the CMM and microscope. The initial position inspected by the microscope must be identical with the edge of profile data, which is difficult to adjust. We adopt a new method to address this problem. Figure 5(c) is the profile of line AA’, and two inspectors are added to assist the confirmation of SSD depth. By the microscope and its screw micrometer, we inspect the surface morphologies along line AA’ and measure the length (signed as D in Fig. 7) from the position (Inspector 1# in Fig. 7) where cracks disappear to the position (Inspector 2# in Fig. 7) where cracks emerge again. Then, we adjust the two inspectors in Fig. 5(c) to be same height and the distance of them [“xDst” in Fig. 5(d)] is equal to D. Move one inspector to the edge of profile (point A or A’), then the height difference of two inspectors [“yDst” in Fig. 5(d)] is considered the SSD depth. Since the SSD at polishing step is localized and can present a large distribution of depths, the measurement of SSD depth should be repeated in different positions on the surface. In this study, the above measurement process is repeated four times along lines AA’, BB’, CC’ and DD’, and their average value is taken as the SSD depth. This operation can eliminate the influence of “localized” SSD, and it can also increase the measurement accuracy of SSD depth.
An experiment for reproducibility of SSD depth measurement is conducted by measuring the SSD depth of the groove in Fig. 5 four times (each time measures the SSD depth along lines AA’, BB’,CC’, DD’, and then average them as the SSD depth of each measurement), with results shown in Table 4. The standard deviation of SSD depth along four lines is 0.064, 0.038, 0.048 and 0.032μm for each measurement, and the standard deviation of four measurements is only 0.026μm, which are quite satisfactory and are believed to be low enough to support the results in this study.
Table 4. Reproducibility Experiment of SSD Depth Measurement
3. Results and discussions
3.1 Surface morphologies after polishing using three pads
The first set of experiments (S1-S3) is conducted to qualitatively inspect SSD characteristics of fused silica before and after polishing using the three pads, with slurry 1# and other conditions listed in Table 5. Before small tool polishing, each segment is etched and inspected under a microscope. Two typical positions, namely A and B, are selected on each segment. With the use of a fixture, the repeated positioning accuracy is less than 10μm to ensure that the microscope can inspect the same position before and after small tool polishing. The three pads then uniformly remove material ~5μm depth; correlative surface morphologies are inspected under a microscope (see Fig. 8, Fig. 9 and Fig. 10).
Table 5. Parameters for Pad Material Investigation
Fig. 8 Surface morphologies of S1 (a) position A before pitch polishing; (b) position B before pitch polishing; (c) position A after pitch polishing; (b) position B after pitch polishing.
Fig. 9 Surface morphologies of S2 (a) position A before polyurethane polishing; (b) position B before polyurethane polishing; (c) position A after polyurethane polishing; (d) position B after polyurethane polishing.
Fig. 10 Surface morphology of S3 (a) position A before leather polishing; (b) position B before leather polishing; (c) position A after leather polishing; (d) position B after leather polishing.
The surface morphologies of S1 at positions A and B prior to pitch polishing are shown in Figs. 8(a) and 8(b). The FOV is 55 × 68μm and one graduation represents 20µm. Numerous trailing indent cracks (<10μm length and ~1μm width) are produced by previous polishing. After pitch tool polishing, the cracks are nearly removed but several cracks and scratches are still found on the surface [Figs. 8(c) and 8(d)]. Therefore, the pitch pad under given conditions is not SSD-free. In critical cases, optics polished by pitch pad may be insufficient and thus post-processing is needed to improve surface morphologies.
The surface morphologies of S2 at positions A and B prior to polyurethane pad polishing are shown in Figs. 9(a) and 9(b). After removing ~5μm material, the cracks are fully removed but some small pits are produced [marked by dotted circles as shown in Figs. 9(c) and 9(d)]. Therefore, polyurethane pad is not considered a SSD-free tool. It exhibits different damage characteristics with pitch pads, and optics polished by polyurethane also need post-processing.
Figures 10(a) and 10(b) are surface morphologies of S3 at positions A and B prior to leather pad polishing. After uniform removal of ~5μm, the surface morphologies are largely improved, and no cracks, pits or scratches can be found in Figs. 10(c) and 10(d). We believe that a leather pad not only can suppress the productions of SSD under given conditions, but also can eliminate the pre-existing SSD developed in previous manufacture process, and it is a promising tool used to polish optics for critical applications. The results in Fig. 10 also indicate that the removal depth of 5μm can ensure the pre-existing SSD after sample preparation in subsection 2.1 can be completely removed, so that not influence the measurement results of SSD depth.
3.2 Surface morphologies after polishing with different slurries
The influence of CeO2 abrasive on SSD characteristics is qualitatively investigated in the second set of experiments (S4-S9). As slurry 1# has been studied, this subsection investigates the performances of slurry 2# and 3# with rough abrasives, with three pads and other conditions as listed in Table 6. Surface morphologies after polishing are recorded in Fig. 11, Fig. 12 and Fig. 13.
Table 6. Parameters for Abrasive Slurry Investigation
Figures 11(a) and 11(b) show surface morphologies of S4 and S5 after polishing by the pitch pad with slurry 2# and 3#, respectively. Numerous scratches and trailing indent cracks are found on the surface. These results, which are much worse than those of Figs. 8(c) and 8(d) by slurry 1#, are mainly caused by rough abrasives of slurries 2# and 3#. Furthermore, these results also indicate that the pitch pad possesses a poor capability to suppress the adverse effect of rough abrasives. In ultra-precision polishing, CeO2 abrasive should has a proper size and narrow size distribution, as well as high purity.
Figures 12(a) and 12(b) are surface morphologies of S6 and S7 after polishing by polyurethane pad with slurry 2# and 3#, respectively. The results are slightly better than pitch pad, but markedly worse than those with the use of slurry 1# [Figs. 9(c) and 9(d)]. Similar to pitch pad, polyurethane pad is also a poor tool for suppressing the adverse effect of rough abrasives.
Figures 13(a) and 13(b) are surface morphologies of S8 and S9 after polishing by leather pad with slurry 2# and 3#, respectively, exhibiting several slight cracks and scratches as illustrated by dotted circles. They are much better than those of pitch and polyurethane pads with slurries 2# and 3#, and are slightly inferior to those of leather pad with slurry 1# [Figs. 10(c) and 10(d)]. These results indicate that leather pads exhibit excellent resistant capability for suppressing and mitigating the adverse effect of rough abrasives. A possible reason for this phenomenon is inferred that a rough abrasive would be mostly pressed into capillary interstices of a leather pad and then the cracks and scratches induced by rough abrasives are sharply mitigated and reduced.
3.3 Surface morphologies after polishing with various pressures and velocities
Pressure and velocity are qualitatively investigated in the third set of experiments (S10-S13) for leather pad with slurry 1#, and other conditions as listed in Table 7. S10, S3 and S11 investigate the influence of pressure with 0.05MPa, 0.10MPa and 0.15MPa, respectively. The surface morphologies, which are clean and crack-free, after polishing with three pressures are shown in Fig. 14. S12, S3 and S13 investigate the influence of spinning velocity with 100rpm, 300rpm and 500rpm, respectively. Surface morphologies presented in Fig. 15 are also SSD-free. These results in Fig. 14 and Fig. 15 prove that pressure and velocity are generally free to SSD, for leather pad with slurry 1#.
Table 7. Parameters for Pressure and Velocity Investigations
Fig. 14 The surface morphologies versus pressure with leather pad and slurry 1#: (a) S10: pressure 0.05MPa; (b) S3: pressure 0.10MPa; (c) S11: pressure 0.15MPa.
Fig. 15 The surface morphologies versus spinning velocity of leather pad with slurry 1#: (a) S12: velocity 100rpm; (b) S3: velocity 300rpm; (c) S13: velocity 500rpm.
3.4 Measurement results and analysis of SSD depth
Based on the qualitative inspections of surface morphologies polished by three pads and three slurries, we find that the SSD depth of fused silica polished by slurry 1# or leather pad is difficult to measure because the crack density on the surface is relatively low. Hence, the SSD depth of fused silica polished by pitch and polyurethane pads with slurry 2# and 3# are quantitatively measured (S4-S7), as well as other six surfaces polished by pitch pad with different pressures, velocities and concentrations (S14-S19). The parameters and results are summarized in Table 8. Thereinto, during each SSD measurement, the standard deviations of SSD depth along four lines are listed in the last column, which are believed low enough to support the following conclusions.
Table 8. SSD Depth Polished by Pitch and Polyurethane Pads
With the same slurry (S4 vs. S6, S5 vs. S7), the polyurethane pad yields smaller SSD depth than the pitch pad, because of its relatively soft working surface. The conclusion is quite in accordance with the results in Fig. 11 and Fig. 12.
With same pad material (S4 vs. S5, S6 vs. S7), slurry 3# yield larger SSD depth as it has more rough particles. The ratios of SSD depth between slurry 3# and slurry 2# are 1.55 and 1.63 for pitch and polyurethane, respectively, which are generally identical with the ratio of abrasive size (D90) of slurry 3# and slurry 2#, that is 1.44. This result is quite in line with most studies about abrasive size on SSD, that a rougher abrasive suffers a larger pressure then induces a deeper SSD. Moreover, compared with the inequalities proposed by Lambropoulos [18] for SSD depth developed in grinding, microgrinding and loose abrasive lapping, the scatter data (all SSD results of Table 8) shown in Fig. 16 suggest the SSD depth of fused silica polished by pitch and polyurethane pads locates nearby the low boundary of Eq. (1), and three of these data for slurry 2# even locate below the low boundary.
Fig. 16 The comparison between SSD depth of fused silica polished by small tools and the law presented by Lambropoulos [18].
With respect to the influence of pressure on SSD depth, the results for S14, S4 and S15 indicate that the increase in pressure (0.05MPa to 0.15MPa) slightly increase the SSD depth (0.91μm to 1.05μm). As the difference of these SSD depths is so small, that it may be caused by the measurement accuracy. However, it absolutely confirms that the pressure is a weaker factor on SSD depth compared with pad material and abrasive size. Ignoring the influence of measurement accuracy, these results indicate that the increase in pressure on a polishing pad not only increases the number of active abrasives, but also slightly increases the maximal load on a single abrasive. The increase of active abrasives increases the material removal rate, and the small increase of load on a single abrasive slightly increases the SSD depth.
SSD results of S16, S4 and S17 show that the increase of spinning velocity (100rpm to 500rpm) slightly reduces the SSD depth. This agrees with the result of Neauport et al. [27] for loose abrasive lapping. Ignoring measurement accuracy, we infer that the increase of spinning velocity largely increase the indentation frequency of abrasives on optical surfaces, which raises removal rate but dulls abrasives more quickly (i.e., degrades the sharpness angle of abrasives). Consequently, the SSD depth is slightly decreased.
SSD results of S18, S19 and S4 indicate that the increase in slurry concentration (5wt% to 12wt%) slightly reduces SSD depth. This is mainly because the increase in slurry concentration increases the number of active abrasives and then reduces the maximal pressure on a single abrasive.
The extensive existence of SSD suggests that the mechanical removal plays one important role in material removal by pitch and polyurethane with CeO2 2# and 3#. Moreover, the SSD depth is ~1/7-1/5 of abrasive size (D90), which is much less than the ratio in grinding and lapping. This is because (1) the pad material is much softer than diamond wheel or cast icon pad, then the pressure on a single abrasive is much smaller; (2) the ductile mode removal, the chemical reaction and surface flow mitigate SSD depth.
3.5 Remove rate and surface roughness
The above experiments have validated that the leather pad has higher SSD resistance, but it does not mean the leather pad is suitable for small tool polishing, unless its material removal rate and surface roughness are comparable to traditional pads. In this subsection, three spot experiments are performed on a Φ150mm fused silica to show the material removal rate and surface roughness of three pads, with orbital radius 9mm, slurry 1#, pressure 0.10MPa, spinning velocity 300rpm, orbital velocity −60rpm and polishing time 2min. A 4 in. (~100mm) Zygo GPI laser interferometer and Wyko NT1100 interferometer are then used to determine the material removal rate and surface roughness, respectively.
The removal map of three pads is shown in Fig. 17, from which the peak removal rate (PRR) and volume removal rate (VRR) of three pads are extracted as shown in Fig. 18, which indicates that the removal rate (both PRR and VRR) of leather pad is comparable with that of polyurethane, and is much higher than that of pitch. This is because leather pad is napped and highly bibulous, and it has no large holes or wide grooves, thus, it has larger contact area with surfaces, which increases the shearing effect in a unit time. The results of surface roughness on the removal spots of the three pads are summarized in Fig. 18. The roughness (Ra values) of fused silica polished by pitch, polyurethane and leather pads are 1.03, 1.64, and 1.13 nm, respectively. Thus, the leather pad shows much better performance than polyurethane on surface roughness, and it is nearly comparable with that for pitch pad. The results can validate the leather pad not only exhibits better resistant capability to SSD, but also has comparable material removal rate and surface roughness to traditional pads, which is a promising pad material used for small tool polishing.
4. Conclusions
This paper investigates subsurface damages (SSDs) of fused silica developed during small tool polishing process. Four key variables, pad material, abrasive slurry, pressure and velocity are studied in detail. The following conclusions could be drawn:
- (1) The polishing leather pad is a nearly SSD-free tool which can be used in fine-polishing and some critical cases. By contrast, pitch and polyurethane pads exhibit poor performance on SSD.
- (2) Rough abrasives can produce numerous cracks, scratches and pits on fused silica. The abrasive should possess a proper size, a narrow size distribution, and high purity.
- (3) Polishing leather pad exhibits excellent capability in suppressing and mitigating the adverse effect of rough abrasives and other contaminations.
- (4) The SSD depths of fused silica polished by the pitch and polyurethane pads (with rough abrasives) are measured by MRF tapper method, which is ~0.77-1.49μm.
- (5) Large pressure, low velocity and low concentration are more likely to yield deeper SSD.
- (6) The removal rate of polishing leather pad is comparable with that of polyurethane pad, and is much larger than that of pitch pad. The pitch pad has optimal surface roughness Ra = 1.03nm, leather pad is nearly same with pitch, Ra = 1.13nm, polyurethane has worse roughness Ra = 1.64nm.
- (7) Polishing leather exhibits excellent SSD performance and large material removal rate, as well as superb roughness. If a flexible interlayer is used to increase its flowability, it can better conform to curved surfaces and is a promising pad for small tool polishing.
Acknowledgments
This study is supported by the National Natural Science Foundation of China (Grant Nos.: 61308075, 61222506), Specialized Research Fund for the Doctoral Program of Higher Education (Grant No.: 20131101110026).
References and links
1. R. A. Jones, “Optimization of computer controlled polishing,” Appl. Opt. 16(1), 218–224 (1977). [CrossRef] [PubMed]
2. H. B. Cheng, Z. J. Feng, K. Cheng, and Y. W. Wang, “Design of a six-axis high precision machine tool and its application in machining aspherical optical mirrors,” Int. J. Mach. Tools Manuf. 45(9), 1085–1094 (2005). [CrossRef]
3. D. W. Kim, W. H. Park, S. W. Kim, and J. H. Burge, “Parametric modeling of edge effects for polishing tool influence functions,” Opt. Express 17(7), 5656–5665 (2009). [CrossRef] [PubMed]
4. D. W. Kim, W. H. Park, H. K. An, and J. H. Burge, “Parametric smoothing model for visco-elastic polishing tools,” Opt. Express 18(21), 22515–22526 (2010). [CrossRef] [PubMed]
5. A. B. Shorey, S. D. Jacobs, W. I. Kordonski, and R. F. Gans, “Experiments and observations regarding the mechanisms of glass removal in magnetorheological finishing,” Appl. Opt. 40(1), 20–33 (2001). [CrossRef] [PubMed]
6. J. C. Lambropoulos, C. L. Miao, and S. D. Jacobs, “Magnetic field effects on shear and normal stresses in magnetorheological finishing,” Opt. Express 18(19), 19713–19723 (2010). [CrossRef] [PubMed]
7. P. M. Shanbhag, M. R. Feinberg, G. Sandri, M. N. Horenstein, and T. G. Bifano, “Ion-beam machining of millimeter scale optics,” Appl. Opt. 39(4), 599–611 (2000). [CrossRef] [PubMed]
8. W. Kordonski, A. Shorey, and A. Sekeres, “New magnetically assisted finishing method: material removal with magnetorheological fluid jet,” Proc. SPIE 5l80, 107–114 (2004). [CrossRef]
9. D. D. Walker, D. Brooks, A. King, R. Freeman, R. Morton, G. McCavana, and S. W. Kim, “The ‘Precessions’ tooling for polishing and figuring flat, spherical and aspheric surfaces,” Opt. Express 11(8), 958–964 (2003). [CrossRef] [PubMed]
10. P. Dumas, C. Hall, B. Hallock, and M. Tricard, “Complete sub-aperture pre-polishing & finishing solution to improve speed and determinism in asphere manufacture,” Proc. SPIE 6671, 667111 (2007). [CrossRef]
11. D. Walker, A. Beaucamp, R. Evans, T. Fox-Leonard, N. Fairhurst, C. Gray, S. Hamidi, H. Li, W. Messelink, J. Mitchell, P. Rees, and G. Yu, “Edge-control and surface-smoothness in sub-aperture polishing of mirror segments,” Proc. SPIE 8450, 84502A (2012). [CrossRef]
12. F. Y. Génin, A. Salleo, T. V. Pistor, and L. L. Chase, “Role of light intensification by cracks in optical breakdown on surfaces,” J. Opt. Soc. Am. A 18(10), 2607–2616 (2001). [CrossRef] [PubMed]
13. M. D. Feit and A. M. Rubenchik, “Influence of subsurface cracks on laser induced surface damage,” Proc. SPIE 5273, 264–272 (2004). [CrossRef]
14. J. H. Campbell, R. A. Hawley-Fedder, C. J. Stolz, J. A. Menapace, M. R. Borden, P. K. Whitman, J. Yu, M. J. Runkel, M. O. Riley, M. D. Feit, and R. P. Hackel, “NIF optical materials and fabrication technologies: an overview,” Proc. SPIE 5341, 84–101 (2004). [CrossRef]
15. M. Buijs and K. K. Houten, “A model for lapping of glass,” J. Mater. Sci. 28(11), 3014–3020 (1993). [CrossRef]
16. J. C. Lambropoulos, Y. Li, P. Funkenbusch, and J. Ruckman, “Non-contact estimate of grinding subsurface damage,” Proc. SPIE 3782, 41–50 (1999). [CrossRef]
17. J. C. Lambropoulos, S. D. Jacobs, and J. Ruckman, “Material removal mechanisms from grinding to polishing,” Ceram. Trans. 102, 113–128 (1999).
18. J. C. Lambropoulos, “From abrasive size to subsurface damage in grinding,” Optical Fabrication and Testing, OSA Technical Digest17–18 (2000).
19. J. A. Randi, J. C. Lambropoulos, and S. D. Jacobs, “Subsurface damage in some single crystalline optical materials,” Appl. Opt. 44(12), 2241–2249 (2005). [CrossRef] [PubMed]
20. S. N. Shafrir, J. C. Lambropoulos, and S. D. Jacobs, “Subsurface damage and microstructure development in precision microground hard ceramics using magnetorheological finishing spots,” Appl. Opt. 46(22), 5500–5515 (2007). [CrossRef] [PubMed]
21. X. Tonnellier, P. Morantz, P. Shore, A. Baldwin, R. Evans, and D. D. Walker, “Subsurface damage in precision ground ULE and Zerodur® surfaces,” Opt. Express 15(19), 12197–12205 (2007). [CrossRef] [PubMed]
22. T. Suratwala, L. Wong, P. Miller, M. D. Feit, J. Menapace, R. Steele, P. Davis, and D. Walmer, “Sub-surface mechanical damage distributions during grinding of fused silica,” J. Non-Cryst. Solids 352(52-54), 5601–5617 (2006). [CrossRef]
23. T. Suratwala, R. Steele, M. D. Feit, L. Wong, P. Miller, J. Menapace, and P. Davis, “Effect of rogue particles on the sub-surface damage of fused silica during grinding/polishing,” J. Non-Cryst. Solids 354(18), 2023–2037 (2008). [CrossRef]
24. P. E. Miller, T. I. Suratwala, L. L. Wong, M. D. Feit, J. A. Menapace, P. J. Davis, and R. A. Steele, “The distribution of subsurface damage in fused silica,” Proc. SPIE 5991, 599101 (2005). [CrossRef]
25. P. P. Hed and D. F. Edwards, “Optical glass fabrication technology. 2: Relationship between surface roughness and subsurface damage,” Appl. Opt. 26(21), 4677–4680 (1987). [CrossRef] [PubMed]
26. D. Golini and S. D. Jacobs, “Physics of loose abrasive microgrinding,” Appl. Opt. 30(19), 2761–2777 (1991). [CrossRef] [PubMed]
27. J. Neauport, J. Destribats, C. Maunier, C. Ambard, P. Cormont, B. Pintault, and O. Rondeau, “Loose abrasive slurries for optical glass lapping,” Appl. Opt. 49(30), 5736–5745 (2010). [CrossRef] [PubMed]
28. J. Neauport, C. Ambard, P. Cormont, N. Darbois, J. Destribats, C. Luitot, and O. Rondeau, “Subsurface damage measurement of ground fused silica parts by HF etching techniques,” Opt. Express 17(22), 20448–20456 (2009). [CrossRef] [PubMed]
29. J. Neauport, P. Cormont, P. Legros, C. Ambard, and J. Destribats, “Imaging subsurface damage of grinded fused silica optics by confocal fluorescence microscopy,” Opt. Express 17(5), 3543–3554 (2009). [CrossRef] [PubMed]
30. R. Laheurte, P. Darnis, N. Darbois, O. Cahuc, and J. Neauport, “Subsurface damage distribution characterization of ground surfaces using Abbott-Firestone curves,” Opt. Express 20(12), 13551–13559 (2012). [CrossRef] [PubMed]
31. P. Blaineau, R. Laheurte, P. Darnis, N. Darbois, O. Cahuc, and J. Neauport, “Relations between subsurface damage depth and surface roughness of grinded fused silica,” Opt. Express 21(25), 30433–30443 (2013). [CrossRef] [PubMed]
32. J. B. Johnson, D. W. Kim, R. E. Parks, and J. H. Burge, “New approach for pre-polish grinding with low subsurface damage,” Proc. SPIE 8126, 81261E (2011). [CrossRef]
33. C. J. Evans, E. Paul, D. Dornfeld, D. A. Lucca, G. Byrne, M. Tricard, F. Klocke, O. Dambon, and B. A. Mullany, “Material removal mechanisms in lapping and polishing,” CIRP Annals-Manufacturing Technology 52(2), 611–633 (2003). [CrossRef]
34. Z. C. Dong, H. B. Cheng, and H. Y. Tam, “Modified subaperture tool influence functions of a flat-pitch polisher with reverse-calculated material removal rate,” Appl. Opt. 53(11), 2455–2464 (2014). [CrossRef] [PubMed]
35. J. E. DeGroote, S. D. Jacobs, L. L. Gregg, A. E. Marino, and J. C. Hayes, “Quantitative characterization of optical polishing pitch,” Proc. SPIE 4451, 209–221 (2001). [CrossRef]
36. Y. G. Li, J. Hou, Q. Xu, J. Wang, W. Yang, and Y. B. Guo, “The characteristics of optics polished with a polyurethane pad,” Opt. Express 16(14), 10285–10293 (2008). [CrossRef] [PubMed]
37. R. Williamson, Field Guide to Optical Fabrication (SPIE, 2011), Chap. 2.
