Surface roughness instrument pdf

The pinhole's effect on the received light is illustrated in the figure on the right. With the laser confocal, the reflected light intesntiy peaks at the focal point. On the other hand, the normal optical system results in a gently sloping curve.

The lack of a peak at the focal point makes it difficult to detect when the target is in focus. For non-contact systems, the light beam spot corresponds to the stylus of contact-systems. Non-contact systems do not directly touch the target, therefore they do not have the disadvantages of stylus wear and the risk of scratching a sample. The size of the beam spot diameter is important in order to accurately measure the profile of a sample.

Generally, the smaller the beam spot diameter, the smaller the features that can be measured. Laser microscopes use lasers for their light sources, which makes it possible to create an extremely small beam spot. When using a x N. A laser microscope can measure asperity with a very small width, which cannot be measured with contact-type systems.

In addition to the points that have already been explained, the characteristics of non-contact types can be summarized as shown below. With a 3D laser scanning microscope, it takes approximately 0. It takes approximately 10 seconds to acquire 3D data with steps in the Z direction. For a single horizontal line, the data for steps can be acquired in approximately 1.

This reflected laser light intensity information can be used to acquire a laser image of the target's surface status. The VK-X Series is equipped with a color CCD camera in addition to the light-receiving element for the laser, which makes it possible to obtain extremely high definition images in color. Because images are created by layering the parts that are in focus, it is possible to acquire observational images that rival those of scanning electron microscopes SEMs.

Results from microscale friction experiments where friction force is proportional to real area of contact on polished and etched silicon surfaces are presented to verify the analysis. A versatile surface processing method based on electrostatic deposition of particles and subsequent dry etching was shown to be able to independently tailor the amplitude and spatial roughness parameters of the resulting surfaces. Statistical models were developed to connect process variables to the amplitude roughness parameters center line average, root mean square and the spatial parameter, autocorrelation length of the final surfaces.

The autocorrelation length of the final surface closely followed a power law decay with particle coverage, the most significant processing parameter. Center line average, root mean square followed a nonlinear relation with particle coverage and particle size. Experimental results on silicon substrates agreed reasonably well with model predictions. This same hybrid surface engineering process was used to demonstrate adhesion and friction reduction.

Microscale adhesion and friction tests were conducted on flat smooth and processed silicon surfaces with a low elastic modulus thermoplastic rubber Santoprene probe that allowed a large enough contact area to observe the feature size effect.

Both adhesion and friction force of the processed surfaces were reduced comparing to that of the flat surfaces. The process is also used to generate superhydrophobic engineering surfaces by mimicking the structure of lotus leaves.

Tunable bimodal roughness in both micro and nano scale and a thin hydrophobic fluorocarbon film were generated on an engineering material surface by the hybrid process.

A geometric model was developed to related air-trapping ability of hydrophobic surfaces with hillock features to process variables hillock diameter, etching depth and coverage and contact angle. The model is shown to be able to predict minimum coverage of hillocks required for air-trapping on hydrophobic rough surfaces. The model predictions agree with experimental observations reasonably well.

This model can particularly be extended to utilizing statistical roughness parameters to predict air-trapping for rough hydrophobic surfaces. There are three common material states, solid, liquid and gas. Theoretically, a material surface is comprised of just several outermost atomic layers of the bulk. Solid surface properties can be categorized as both physical and chemical properties.

Physical properties are those properties that do not depend on atomic elements, for example, surface morphology, which only depends on atomic arrangement. On the other hand, surface chemical properties depend on the elements, for example, surface energy, which is higher for silicon than that for wax.

Due to unbalanced atomic forces, surface atoms have more energy than bulk atoms, and react more readily with the environment. Atoms in surface layers are relaxed and arranged differently from the bulk to minimize energy.

In addition, contaminations can also cause changes in surface morphology. In ambient conditions, atoms in surface layers are always physically and chemically changed by environment, for example, through oxidation and adsorption and hence experiments may often be conducted under high vacuum.

In manufacturing processes, surface morphology can be affected by many sources, for example, inaccuracies in the machine tools, deformation under cutting forces, vibrations of the machine or workpiece, which will lead to irregularity in millimeter scale; and rupture of the material during the chip removal, which will cause micro scale irregularity. In other words, surface morphology is scale dependent. In many engineering applications, it is convenient to categorize surface morphology as surface roughness, waviness and lay at the micrometer, millimeter and meter scale respectively, as shown in Fig.

In the nanoscale nanometer scale , surface roughness is associated with atomic structure, where the location of a single atom is important. Research areas in which roughness plays a prominent role range from quality assurance, tribology friction, lubrication and wear , biomechanics, hydrodynamics, to oceanography and selenology1,2.

This thesis will focus on surface roughness of solid materials and its effects on solid- solid and solid-liquid interactions in ambient conditions. First, surface roughness is described using statistical roughness parameters and the effect of a spatial roughness parameter, autocorrelation length on contact and friction behavior is analyzed, via a statistical contact Figure 1. Then, a novel hybrid surface processing method that combined electrostatic deposition and plasma etching is described that can generate random surfaces with desired spatial and amplitude roughness parameters.

This process is shown to generate surfaces with reduced adhesion and friction. Finally, the hybrid processing method and statistical modeling of surface morphology was used in the study of hydrophobic rough surface wetting. Surface roughness modeling Roughness measurement Roughness modeling starts from the measurement of height variation of profiles one- dimension, 1D or surfaces two-dimension, 2D. Generally the measured height data are in discrete format and the distance between two data points sampling distance is constant.

Profiles are easier to measure and many surfaces are generated from parallel profiles. Stylus instruments are widely used in profile measurements. The basic principle of these instruments is similar: a probe scans along a profile and its interactions with the substrate are recorded and analyzed to obtain profile height information.

The interactions may be contact forces for many stylus instruments and atomic force microscopy; or tunneling current for scanning tunneling microscopy.

Resolution of profile height measurement depends on probe size, which is usually in the micron- millimeter scale for stylus instruments and in the nano- to microscale for SPM.

There are also some other measuring methods, for example, optical methods, which may apply geometrical optics or physical optics; contact methods, electrical methods, and fluid methods2. Due to the finite nature of the probe size, it may not be able to follow irregularities that are separated by a distance smaller than the probe size. This lower limit may also be introduced when analog signals are sampled into digital signals. On the other hand, the total length of measurement is the large wavelength cut-off.

This means that the measured irregularities are actually sampled signals from a profile by a band pass filter, whose higher frequency limit and lower frequency limit are decided by instrument and scanning length respectively.

The appropriate selection of the two limits should be based on the application, for example, friction study of a large industrial bearing may only need a high frequency limit of surface roughness at the micron scale while friction study of a microfabricated bearing requires a much larger high frequency limit of surface roughness, which may be beyond the capability of a common stylus instrument. Roughness parameters The modeling of profile roughness starts from the definition of a mean line, which is parallel to the geometrical profile such that the area of solid above it is equal to the area of void below it as the dot-dashed line shown in Fig.

Roughness parameters can be divided into average roughness parameters, statistical parameters, random process parameters and fractal parameters. These parameters are calculated only based on measured data set.

If a distribution function of height data can be obtained, for example, estimated from the histogram of measured data, statistical roughness parameters can be defined, such as bearing area curve, skewness and kurtosis. Both skewness and kurtosis are measures of inefficiency in a statistical sense, which means their values are sensitive to both effects of surface properties and sampling artifact. Further, if the sequence of measured data points is considered i. A 1D profile with its mean line shown as the dot-dashed line.

Since roughness is scale dependent, roughness parameters based on measured data are also scale dependent. This makes it impossible to compare roughness parameters without knowing scale information, for example, sampling distance. Fractal methods provide scale- independent roughness parameters such as the fractal dimension and Hurst exponent, by assuming surface morphology is self-affine3.

However this assumption is not true because engineering surfaces cannot be self-affine in all scales4. Roughness parameters can also be divided into amplitude parameters and spatial parameters based on what information they can provide.

Spatial parameters provide lateral information, for example, peak density, zero crossing density and ACL. Compared to amplitude parameters, the effect of spatial parameters on engineering applications, for example, tribology, is not obvious, thus they are not well studied. The study of spatial parameters are often based on correlation function and sampling distance.

Longuet-Higgins studied a random, moving surface, which was assumed to obey Gaussian distribution and be isotropic. There is no limitation on the form of correlation function, but the energy spectrum was assumed to be circular symmetry about the origin. Furthermore, if the spectrum has predominantly one wavelength, a number of statistical properties of the random surface, for example, the average density of maxima per unit area, can be obtained as functions of the wavelength in simplified forms5.

Whitehouse and Philips used discrete random process analysis to express tribological parameters in terms of just two points on the measured autocorrelation function and the RMS value of the surface with the relaxation that autocorrelation function can have a general form7.

Whitehouse, Philips extended their analysis to 2-D surfaces which can be expressed in terms of between four and seven points on the autocorrelation function depending on the type of surfaces8. The nature and magnitude of the difference between results from the discrete analysis and those from continuous theory was discussed in details.

Greenwood predicted the properties of the peaks and summits of a rough surface based on the assumption that the surface is two-dimensional random noise and found the predictions in non-dimensional form dependent only to a minor degree on the surface parameters while the absolute values are strongly dependent on the sampling interval9.

Contact models Solid-solid contact is very common in mechanical and electrical systems, for example, rolling bearing, and electrical switch. For rough surfaces, the real area of contact is much smaller than the nominal area of contact as shown in Fig. The reduction of contact area affects friction, adhesion, wear and other tribological phenomena as well as electrical conductivity.

The real area of contact is not only a function of surface roughness, but also a function of applied normal load. For elastic contact, real area of contact increases with an increase of normal load. The real area of contact is much smaller than the nominal contact area. One early developed model of contact mechanics is the Hertzian model, which is restricted to frictionless surfaces and perfectly elastic solids Based on Hertzian mechanics, several models have been developed by releasing its restrictions, for example, the inclusion of lateral forces and the introduction of plasticity and viscoelasticity.

During the study of adhesion forces between rubber and substrates, surface energy was introduced to explain the finite size of contact area at zero normal load, which is the Johnson-Kendall-Roberts JKR theory As pointed by Tabor, JKR theory ignored attractive forces outside the contact area.

Derjaguin, Muller and Toporov developed a contact model, which assumed that molecular forces act in a ring-shaped zone around the contact area and decrease quickly The shortcoming in DMT theory is the neglect of the deformation due to attractive forces close to the edge of the contact area.

By analogy to fracture mechanics, Maguis et al. For contact of two rough surfaces, it was shown that this case can always be approximated to the contact of a flat surface with a rough surface with small errors. From contact mechanics models, the relation between normal load and contact area of a smooth sphere and a flat surface could be obtained.

The summation of normal load and contact area for every single sphere forms the macro load and macro real area of contact.

Depending on the definition of surface peaks and treatment of peaks distribution and shape, there are lots of models to describe the macro relation between load and real contact area. Greenwood and Williamson assumed a statistical amplitude distribution function Gaussian, exponential or others of surface peaks and a constant radius of peak curvature Thus the total real area of contact at a given intersection of the rough surface into the flat surface can be calculated from the peak amplitude distribution.

For Gaussian amplitude distribution, the density function depends on the RMS of surface roughness. After that, more complex models have been developed based on the Greenwood-Williamson GW theory by treating peaks as ellipsoids, or by introducing a distribution of peak size3. Whitehouse and Archard assumed a normal distribution and an exponential form ACF of surface heights. Then using 3-points peak definition, the amplitude distribution function of surface peaks can be written as a function of ACL.

Following the same procedure of GW theory, the real area of contact can be derived. In the Whitehouse and Archard theory, curvature of peaks is a variable6. Nayak assumed that the surface heights, slopes and curvatures possess a multi-Gaussian probability density and derived the statistical peak distribution based on surface amplitude moments All the above models assumed isotropic surface morphology. For anisotropic Gaussian surfaces, Longuet-Higgins5 developed a general description and the corresponding anisotropic surface contact model based on this general description is very complicated.

In the following, we will only consider the isotropic surfaces. From the above literature review on surface roughness modeling, it is clear that amplitude roughness parameters, particularly RMS, are widely used and accepted in surface characterization and tribological study.

Thus it is worthwhile to develop a simplified contact model that can reveal the importance of ACL on surface characterization and engineering applications. Experimental and analytical models have been developed to predict surface roughness for some manufacturing processes, for example hard turning and grinding There is also classification of manufacturing processes based on ranges of surface roughness parameters that result for example, RMS, as shown in Fig.

Surface roughness could be formed during removing materials, adding materials or rearrange materials. Here we will review some surface processing techniques according to the three categories. Material removing processes Grinding, lathing, laser machining and top-down microfabrication techniques generate surface by removing materials.

There are many sources that contribute to surface roughness formation during material removing, for example, rupture of materials, vibration or random errors during processing. Because particle size and shape are random and material remove by a single particle is also random, the generated surface morphology is a summation of many independent random variables. From Central Limit theory, we know that height distribution of ground surfaces should follow Gaussian function. By using ideal conic Figure 4.

Classification of processing methods based on CLA range of generated surfaces. But it is hard to tune those parameters in order to generated random surfaces with desired roughness parameters. Laser micromachining is another powerful surface processing techniques, which could be applied to glass, ceramic, titanium and other materials that are hard to process with traditional machining techniques, like grinding. The interaction between laser beam and material is complex and we will only consider thermal effects.

By adsorbing laser energy, material around focus point is heated up and evaporated. There will be debris formed from melted material or sputtering. Thermal shock and other thermal effects may cause surface morphology change too. Thus generated surface roughness depends on both material and laser properties and is hard to control. Laser micromachining is a sequential process and not suitable for large surface processing. Chemical and physical etching is widely used in the semiconductor industry as well as in fabrication of microelectronic mechanical systems MEMS.

Periodic patterns are different from surface roughness because they are not random and they are designed to achieve certain functions.. Because the selection of surface roughness scale depends on the application, sometimes microfabricated periodic patterns are still treated as surface roughness for macro applications, for example the study of texture effects on lubrication.

The nanoscale or atomic scale roughness could be generated by removing atoms one by one. During chemical etching, etchant reacts with atoms on surfaces and the generated new compounds are removed by fluid or vacuum. Thus the arrangement of atoms on surfaces and the exposure of atoms to etchant as well as etchant density could affect the formation and remove of new compounds.

Hydrogen bubbles generated during KOH etching of silicon were believed to cause pyramid formation on etched surface because they can block etchant in certain direction During physical etching, ion or electron bombarded onto a surface and kick out atoms on the surface. The removing process is random and the generated surface is random too. Certainly physical and chemical etch could be combined into one process, for example, reactive ion etch, where physical bombardment of electrons assists the chemical reaction.

Generally it is hard to tune spatial roughness structures because the difficulty to control etchant position. Material adding processes Electroplating, thin film deposition and other bottom-up techniques form surfaces by adding materials.

Atomic flat surface can be generated by well controlled deposition process, such as atomic layer deposition ALD. During ALD, precursors are separated from the substrate throughout the coating process and film growth is self-limited and based on surface reactions. This helps remove pin-hole and avoid grain formation. In many other thin film deposition techniques, such as chemical vapor deposition CVD , or physical vapor deposition PVD , there are so many precursors available that pin- hole or grains are always formed on the substrate.

By controlling temperature, gas supplement rate and other conditions, grain size and surface morphology can be varied. Electroplating is a process that can coat an electrically conductive object with metals using electrical current to improve its abrasion and wear resistance, corrosion protection and other properties. It is also used to deposit solder bumps in flip-chip packaging and surface roughness can be reduced by an addition of a surface active agent But generally the variation of surface morphology by deposition is very limited because deposition processes are slow and lack of precise control in spatial structures.

This step can be repeated to form multi-layer coating as long as the negative charged coating and the positively charged one can be added.

In general, this multi-layer coating process is called layer-by-layer LbL deposition, where other kind of holding forces can be used, for example, hydrogen-bonding Surface morphology of coated surfaces depends on molecule particle shape, defects formation during deposition and interactions among molecules particles Surface morphology can be controlled by varying molecular type, geometry, environment and molecular interactions But electrostatic force is not strong enough to hold large particles, thus the modification of surface morphology may not be very useful in real applications.

Molding techniques Materials may be reorganized to form new surfaces, for example, by heating and cooling, chemical reactions. Inject molding is widely used in industry to obtain plastic and metal parts, where molten materials are injected at high pressure into a mold and separated from the mold after cooling.

Surface morphology of injection molding of thermoplastic polymers generally depends on polymer crystallinity, crystal structure and dimensions controlled by the molding process PDMS micromold techniques utilize polymer solidification to generate solid surfaces. The control of generated surface roughness is difficult. Although there is a large range of processes across scales that can be used to vary the final roughness of a material, in general the level of control is very small.

In addition, there appears to be a lack of ability to tune both amplitude and spatial roughness parameters independently and precisely. Liquid may form a droplet or spread on the solid surface depending on the interfacial properties, such as solid surface roughness, surface energy and liquid surface tension.

Currently understood effects of surface roughness on the formation of droplet on a solid surface will be described next. Superhydrophobicity Contact angle is the primary measurement to determine wettability, which measures the angle between the solid surface and the surface of a liquid droplet on the surface, as shown in figure 6. Superhydrophobic surfaces are found in nature on the leaf surfaces of many plants such as the lotus and colrabi These surfaces also exhibit self-cleaning capability by which rolling water droplets remove dirt and debris from their surfaces.

Figure 6. Illustration of contact angle. Considering that surface roughness increases the interfacial area between solid and liquid, Wenzel proposed a correction factor r for contact angle on rough surfaces, which is equal to the ratio of rough interfacial area over flat interfacial area under the droplet And they did not provide a criterion to predict whether air-trapping could be formed.

Thus, it is important and necessary to develop a criterion that can evaluate air-trapping ability of hydrophobic rough surfaces and correct contact angle based on common roughness parameters, such as RMS, ACL, etc..

Several methods have been employed to generate engineering surfaces that can mimic the structure and chemistry 24 of natural superhydrophobic surfaces. Polymer coatings or layer-by-layer deposited particles with both low surface energy and micro structures can be attached to the bulk to achieve superhydrophobic properties Microfabrication techniques are widely used to generate predetermined micro roughness while aligned carbon nanotubes have been utilized 31,32 for nano scale roughness.

Since many engineering materials are hydrophilic, a further step is needed to coat the roughened substrates with a hydrophobic layer self assembled 33 monolayers , polymer films34 or diamond-like carbon films It is of interest to develop processing methods that allow a high degree of control over the resulting surface structure and that can simultaneously impart hydrophobicity to render a material superhydrophobic.

The overall research plan to achieve these objectives is shown in Fig. The rest of this thesis is organized according to the research activities performed as outlined below.

Overall structures of research plan. A simplified statistical contact model will be introduced and confirmed with experimental data. Chapter 3 will discuss a hybrid surface processing method based on electrostatic deposition of colloidal particles and plasma etching, which can tune autocorrelation length of generated surfaces.

Special attention was given to generate random patterns based on random deposition of colloidal particles. Generated surfaces were measured to model roughness parameters using process variables.

Chapter 4 will show that the developed surface processing method can tune surface amplitude roughness parameters, such as RMS, CLA as well. This process was used to generate surfaces with reduced adhesion and friction as discussed in Chapter 5. This process was also used to achieve superhydrophobic surfaces for objective c as described in Chapter 6. Based on geometrical requirement for air-trapping formation, relations between roughness parameters and superhydrophobicity were also developed as discussed in Chapter 7.

Finally Chapter 8 will cover the conclusions and future work. Appendix B is the matlab codes used in Chapter 3. References 1 B. Greenwood and J. Wu, Meccanica 36, Whitehouse, Wear , Whitehouse and J. Whitehouse and M. Maugis, Contact, adhesion, and rupture of elastic solids springer, Berlin, Nayak, wear 26, Bush, R. Gibson, and G. Ozel and Y.

Hecker and S. Campbell, K. Cooper, L. Dixon, R. Earwaker, S. Port, and D. Schiffrin, Journal of Micromechanics and Microengineering 5, Kim, J. Yu, J. Lee, and T. Lee, Journal of Electronic Materials 33, Seo, J. Lutkenhaus, J. Kim, P. Hammond, and K. Char, Macromolecules 40, Pantani, I.

Coccorullo, V. Speranza, and G. Titomanlio, Progress in Polymer Science 30, Barthlott and C. Neinhuis, Planta , Wenzel, Industrial and Engineering Chemistry 28, Cassie and S. Baxter, Transactions of the Faraday Society 40, Feng and L. Jiang, Advanced Materials 18, Ji, J. Fu, and J. Shen, Advanced Materials 18, Wang, L. Feng, and L. Zhao, J. Xu, Q. Xie, L. Weng, X. Guo, X. Zhang, and L. Shi, Macromolecular Rapid Communications 26, Zhang, D. Wang, Z. Gu, and H. Mohwald, Langmuir 21, Xie, G.

Fan, N. Zhao, X. Guo, J. Xu, J. Dong, L. Zhang, Y. Zhang, and C. Nakajima, A. Fujishima, K. Hashimoto, and T. Watanabe, Advanced Materials 11, Zhu, Y. Xiu, J. Xu, P. Tamirisa, D. Hess, and C. Wong, Langmuir 21, Lau, J. Bico, K.

Teo, M. Chhowalla, G. Amaratunga, W. Milne, G. McKinley, and K. Gleason, Nano Letters 3, Genzer and K. Efimenko, Science , Hozumi and O. Takai, Thin Solid Films , Hatada and K. This paper presents a relationship between ACL and the friction behavior of a rough surface. The probability density function of peaks and the mean peak height of a profile are given as functions of its ACL. These results are used to estimate number of contact points when a rough surface comes into contact with a flat surface and it is shown that the larger the ACL of the rough surface, the less the number of contact points.

Based on Hertzian contact mechanics, it is shown that real area of contact increases with increasing of number of contact points. Since adhesive friction force is proportional to real area of contact, this suggests that the adhesive friction behavior of a surface will be inversely proportional to its ACL. Results from microscale friction experiments on polished and etched silicon surfaces are presented to verify the analysis.

If a surface can be described by a Gaussian random process, statistical parameters describing the surface height variation can be completely defined based on its height distribution and autocorrelation function, that is, the surface is totally determined in a statistical sense.

A lot of work has been done to derive surface roughness descriptors that can be practically used for both 1-D profile and 2-D surface situations A Gaussian distribution of ordinates has been experimentally shown to be valid for many engineering surfaces4.

Another common assumption is that the autocorrelation function has an exponential form, which has also been verified experimentally In order to simplify calculation, surfaces are assumed to be isotropic, even though it is believed that at least in principle the same analysis can be extended to anisotropic surfaces5. Special emphasis has been placed on the distribution of peaks because when two surfaces approach each other, the peaks will undergo contact first, and the distribution of peaks will decide the real area of contact.

This reasoning is true especially in predominantly elastic contacts and is utilized in many friction models4,12, It is well known that both height parameters and spatial functions are needed to describe the height distribution and texture of a rough surface Autocorrelation function and autocorrelation length ACL have been widely used in surface related studies to provide spatial information in addition to amplitude parameters, such as Root-Mean-Square roughness. Generally, ACL is used to measure how quickly a random event decays or the distance over which two points can be treated as independent in a random process.

In early friction studies, Rabinowicz had deduced the size of contact junctions based on a simple model of the sliding process using autocorrelation analysis, where the junctions were assumed to have the same size, but different shear strengths However, a detailed analysis of ACL and its relation, if any, to friction behavior is lacking. The main objective of this paper is to analyze the effects of ACL on real area of contact and adhesive friction force based on peak analysis.

A physical understanding of ACL that is suitable for tribological applications is proposed based on statistical description of rough surfaces and Hertzian contact mechanics. Results from microscale friction experiments of silicon samples with different ACL are provided to verify the analysis.

Now we consider profiles from different surfaces that obey a normal distribution and that are sampled at the same interval. We also assume every profile has an exponential autocorrelation function with different ACL values.

Probability of an ordinate being a peak at height y. This makes it possible to compare peak distribution of profiles from different surfaces using only ACL if sampling intervals are kept constant for the comparison. From the figure it is clear that a profile with a larger ACL has a broader peak distribution and the mean peak height is closer to the mean line than a profile with a smaller ACL. It can be seen that as the ACL decreases, the mean peak height moves away from the mean line up to around 0.

From Fig. Since peaks will come into contact first when two surfaces approach each other, the total number of contact points under a given load will be equal to the number of peaks above a determined height if no peaks merge together under the load. Based on the above peak analysis, the relation between number of contact points and ACL is shown in Fig. It is clear that a larger ACL leads to a smaller number of contact points while a smaller ACL leads to a larger number of contact points.

Since all surfaces are isotropic, the above conclusion for 1D profile should still be true for 2D surface in a qualitative sense. Variation in number of contact points and real area of contact with ACL for a rough surface in contact with a flat surface at a given load.

ACL and Real Area of Contact at low loads When two solid surfaces come into contact, surface roughness causes contact to occur at discrete contact spots, and the total area of which is a small fraction of the nominal area of contact. This real area of contact dictates tribological behavior of the interface and is used in many friction and wear models Researchers have also developed methods to experimentally evaluate real area of contact using for example, contact resistance29, ultrasonic30,31 or optical methods However, experimental determination of real area of contact in various systems still remains non-trivial.

Here we employ a Hertzian- mechanics-based model to estimate real area of contact. For simplification, we assume one of the two contacting surfaces is perfectly flat while the other one is rough and its peaks can be treated as spheres with the same radius.

Second, we assume the number of contact points at a total load P to be n. Figure 3 also plots the normalized real area of contact Eq. It can be seen that a surface with a smaller ACL will have a larger real area of contact than a surface with a larger value of ACL.

ACL and Adhesive Friction Force The friction force between two solid surfaces in relative motion against each other is composed of an adhesive component and a deformation component. It has been shown previously in section B that under low loads, a surface with larger ACL would exhibit a smaller real area of contact as compared to a surface with smaller ACL.

Hence, for a given material and relatively low loads, a surface with larger ACL will exhibit a lower adhesive friction response than a surface with a smaller value of ACL. Experiment We performed friction experiments on silicon samples with different surface roughness characteristics to verify our analysis.

All silicon samples were made from n-type silicon wafer from Virginia Semiconductor, Inc. Fredericksburg, Virginia. Wafers were cut to 1. We prepared Si samples to exhibit different levels of roughness via polishing and etching techniques. After the polishing sequence, the samples were ultrasonically rinsed in de-ionized DI water for 10 minutes.

For chemically etched samples, the coupons were first dipped into BOE at room temperature for 1 minute to remove the native oxidized layer. This was followed by a rinse in DI water for 3 minutes.

Finally the samples were rinsed in DI water for 5 minutes. During the etching procedure, samples were kept wet to minimize attachment of particles to sample surfaces. Figure 5. Microtribometer friction force data as a function of normal load for prepared Si samples. Data from seven runs are shown. Coefficient of friction values slope from linear fits of the data are shown. The R2 values of the linear fits are 0.

Since the particles are very small, their effect on the microscale friction measurement described next can be neglected. These particles were not included in the calculation of surface roughness parameters.

The probe was a silicon nitride ball with radius of 1. The normal load was linearly increased from 0 to 15 mN during each stroke. Seven runs for every sample were performed.

Figure 5 shows all the friction data for both samples. Both samples exhibited some stick-slip like variation at higher loads. It can be seen that the polished Si sample exhibits higher friction forces than the KOH-etched sample. Table I: Average roughness parameters for polished and etched samples at 10um scan size obtained from AFM scans.

Table 1 lists the measured roughness parameters of the prepared surfaces. It was originally designed by Dr. It consists of a diamond probe stylus held by spring pressure against the surface of a lapped steel cylinder and is attached to the body of the instrument by a leaf spring. The lapped cylinder is supported on one side by the probe and on the either side by fixed rollers. Alight spring steel arm is attached to the lapped cylinder. It carries at its tip a diamond scriber which rests against a smoked glass.

The motions of the stylus in all the directions except the vertical one are prevented by the forces exerted by the two springs. For measuring surface finish the body of the instrument is moved across the surface by screw rotated by asynchronous motor.

The vertical movement of the probe caused by surface irregularities makes the horizontal lapped cylinder to roll. This causes the movement of the arm attached to the lapped cylinder. A magnified vertical movement of the diamond scriber on smoked glass is obtained by the movement of the arm. This vertical movement of the scriber together with horizontal movement produces a trace on the smoked glass plate. This trace is further magnified at X 50 or X by an optical projector for examination.

Taylor-Hobson Talysurf is a stylus and skid type of instrument working on carrier modulating principle.

Its response is more rapid and accurate as compared to Tomlinson Surface Meter. The measuring head of this instrument consists of a sharply pointed diamond stylus of about 0.

In this instrument the stylus is made to trace the profile of the surface irregularities, and the oscillatory movement of the stylus is converted into changes in electric current by the arrangement as shown in Fig. The arm carrying the stylus forms an armature which pivots about the centre piece of E-shaped stamping. On two legs of outer pole pieces 'the E-shaped stamping there are coils carrying an a.

These two coils with other two resistances form an oscillator. As the armature is pivoted about the central leg, any movement of the stylus causes the air gap to vary and thus the amplitude of the original a.

The output of the bridge thus consists of modulation only as shown in This is further demodulated so that the current now is directly proportional to the vertical displacement of the stylus only.

The work to be tested is placed on the table of the instrument. The work and the table are traversed with the help of a lead screw.

The stylus which is pivoted to a mirror moves over the tested surface. Oscillations of the tracer point are transmitted to the mirror. A light source sends a beam of light through lens and a precision slit to the oscillating mirror. The reflected beam is directed to a revolving drum, upon which a sensitised film is arranged.

This drum is rotated through two bevel gears from the same lead screw that moves the table of the instrument. A profilogram will be obtained from the sensitized film, that may be sub-sequently analyzed to determine the value of the surface roughness. Open navigation menu. Close suggestions Search Search. User Settings.

Skip carousel. Carousel Previous. Carousel Next. What is Scribd? Explore Ebooks. Bestsellers Editors' Picks All Ebooks. Explore Audiobooks. Bestsellers Editors' Picks All audiobooks.

Explore Magazines. Editors' Picks All magazines. Explore Podcasts All podcasts. Difficulty Beginner Intermediate Advanced. Explore Documents. Surface Roughness. Uploaded by Santosh Santu. Document Information click to expand document information Description: Surface roughness -ppt.

Original Title Surface roughness -ppt. Did you find this document useful? Is this content inappropriate? Report this Document. Description: Surface roughness -ppt. Flag for inappropriate content. Save Save Surface roughness -ppt For Later. Original Title: Surface roughness -ppt.

Related titles. Carousel Previous Carousel Next.