Ground Silicon Wafers Research Articles

Effect of micromorphology on measurement of residual stress of ground silicon wafers using Raman spectroscopy

Effect of micromorphology on measurement of residual stress of ground silicon wafers using Raman spectroscopy

Residual stress distribution of silicon wafers machined by rotational grinding based on molecular dynamics

Wafer rotational grinding is widely used to thin silicon wafers and the grinding-induced residual stress distribution is desired to be known. However, it is difficult to obtain it directly by measurement since the variation of residual stress values of different locations tends to be submerged in the fluctuations of measured values. In this study, molecular dynamics was employed to obtain the residual stress states of ground monocrystalline silicon of different crystal directions. The actual cutting directions of the abrasive grains were obtained as the tangential directions of the abrasive grain's traces on the silicon wafer surface relative to the crystal coordinate system. It was found that the actual cutting direction of an abrasive grain along a single grinding trace would change from the center to the edge of the silicon wafer since the grinding traces are curved lines. The residual stress state at a specific point on the wafer surface was obtained by mapping the residual stress values obtained by molecular dynamics to the actual cutting direction. The residual stress distribution map was obtained successfully and any point of the ground silicon wafer could be known. The location of maximum residual stress was found to deviate from the [1 1 0] crystalline direction on the ground wafer due to the change in the actual cutting direction relative to it.

Ductile deformation and subsurface damage evolution mechanism of silicon wafer induced by ultra-precision grinding process

In the field of advanced packaging, large-size silicon wafers are usually thinned by ultra-precision grinding technology relying on the process of workpiece self-rotation. However, the damage caused by mechanical material removal deteriorates the wafer surface flatness and die strength, while its atomic-scale formation principle has yet to be revealed. This work explores the ductile deformation and subsurface damage evolution mechanism of silicon wafer during the ultra-precision grinding process. Some grinding tests are first conducted using different diamond wheels. The surface topography and energy dispersive spectroscopy (EDS) mapping of ground wafers are measured by scanning electron microscope (SEM). The phase transition properties of the ground silicon wafer are confirmed via Raman spectrum. Next, the specimen of groove cross-section is fabricated and its morphology is examined via transmission electron microscopy (TEM) to analyze the subsurface damage characteristics. The subsurface defects at the atomic scale, involving amorphous layer, dislocations, stacking faults and lattice distortions, are observed. Due to its controllable load and similarity of the cutting motion, nanoscratch is utilized for investigating the influence of abrasive interactions on subsurface deformation during the grinding process. Thus, a series of multiple nanoscratch tests are performed in both varied and constant force modes. The material removal behaviors under different scratching conditions are illustrated. This work provides a fundamental understanding of developing a high-efficiency and low-damage thinning method for 12-inch silicon wafers.

Modeling on the Flatness of Silicon Wafers Ground by the Grind-polishing Wheel

The grind-polishing wheel is a new non-woven structure grinding tool with low elastic modulus and ultra-fine abrasive grains, which can obtain high surface/sub-surface quality ground silicon wafers. To obtain high-shape flatness during grinding with the grind-polishing wheel, a prediction model of the ground silicon wafers’ flatness considering the elastic modulus of the wheel, processing parameters, and abrasive grain size is developed. The grinding experiments under different speed ratios were conducted to verify the prediction model accuracy. It shows that the model can successfully predict the flatness of ground silicon wafers, and the error of the prediction model is less than 13%. The proposed model has some guiding significance for choosing the appropriate character of the grind-polishing wheel and grinding parameters to achieve desirable flatness for silicon wafers.

Mechanical effect of abrasives on silicon surface in chemo-mechanical grinding

To obtain high-quality surfaces and subsurfaces of ground silicon wafers, chemo-mechanical grinding (CMG) for silicon has been developed by inducing chemical reactions into grinding. Chemical reactions between abrasives and silicon during CMG process have been proved in previous studies. The mechanical effect of abrasives in CMG process is not fully understood. To investigate abrasives’ mechanical effect on material removal and subsurface damage of silicon in CMG process, silicon nanoscratching experiments with CeO2 indenter under a progressive normal load were conducted, and microtopographies of scratch surface and subsurface were observed. For comparison, silicon nanoscratching experiments with diamond indenter were also conducted. Finite element models of silicon nanoscratching with CeO2 and diamond indenters were built to investigate the relationship between stress distribution and damage evolution of silicon. Experimental investigations and simulation results indicate that no obvious material removal of silicon or serious subsurface damage is generated by CeO2 indenter's mechanical action because the wear of CeO2 indenter leads to the release of the stress on silicon. Material removal of silicon during CMG process relies on chemical reactions between soft abrasives and silicon other than stress on silicon generated by soft abrasives.

Effects of grinding-induced surface topography on the material removal mechanism of silicon chemical mechanical polishing

Ultra-precision thinning technology using workpiece self-rotational grinding followed by chemical mechanical polishing (CMP) is extensively applied in the integrated circuit manufacturing process, which enables to obtain large size ultra-thin silicon wafers with high material removal rate and low damage layer thickness. However, an in-depth understanding of the influence caused by ground surface topography on material removal mechanism in silicon CMP process has not been revealed yet. This work systematically investigates the contact characteristics of the wafer-pad interface and corrosion behaviors of the ground silicon wafer immersed in polishing solution. Firstly, some silicon wafers are ground using different wheels and subsequently polished. The material removal depth during the whole CMP process is measured. Then, the intrinsic model of polishing pad is determined via mechanical property tests. A finite element method is adopted to evaluate the contact status, displacement distortion and stress distribution at the contact region between pad and ground wafer surface. Moreover, the chemical reaction mechanism between ground silicon wafer and polishing solution is revealed by utilizing X-ray photoelectron spectroscopy and electrochemical analysis. The influence of wafer surface topography on corrosion resistance in CMP process is illustrated. Finally, corresponding experimental results are explained from an atomic scale by a ReaxFF reactive molecular dynamics simulation model. This presented study demonstrates the corrosion promotion principle of ground silicon wafer in CMP process.

Multimode photoacoustic characterization of subsurface damage in ground thin wafers

A multimode photoacoustic method has been applied for evaluating the degree of subsurface damage (SSD) in the ground silicon wafer. Mode conversion of laser-generated broadband guided waves is caused by wafer thinning. Bayesian inversion is applied to quantify the effective elastic constants of the SSD layer together with the corresponding uncertainty estimates from the experimental dispersion data of multimode guided waves. Laser ultrasonic experiments have been performed on the polished and ground silicon wafers with different thicknesses to determine the ability of the method to characterize SSD and the effect of acoustic mode conversion on the quantization results. The experimental results show that the mode of the laser-generated broadband guided waves will gradually change from Rayleigh wave to Lamb wave as the wafer is thinned. Both the Rayleigh mode and the A0 Lamb mode can accurately estimate the effective elastic constants of SSD, but the A0 Lamb mode leads to longer confidence intervals, which has been theoretically explained.

A novel dry processing technique to construct a high-performance superfine diamond grinding wheel with thermal and material characterizations

A novel dry process that prepared in mixing with a pore-forming agent employs a single batch process in production of grinding wheels with the advantage without using resin liquid as an additive binder. The agglomeration of a pore-forming agent can be avoided due to no dressing agent mixing with the superfine diamond abrasives and binders. Bulk mass production can still ensure to fulfill the excellent homogeneity in mixing cross the entire grinding wheel. Therefore, an uneven mixing caused by mixing errors and human factors can be minimized. The grinding wheel can reach an ultrahigh porosity (~82 %) at a lower sintering temperature of 620 °C. Low-temperature sintering can effectively reduce the carbon footprint. A thermogravimetric-differential scanning calorimeter (TG-DSC) was used to determine the thermal properties of superfine frits. Further characterization of ultrafine binders by an optical non-contact dilatometry is utilized to set-up a protocol sintering plan of grinding wheels. Microstructure from scanning electron microscope (SEM) measurement on grinding wheel presents a uniform large pore structure which offers a benefit to wafer surface removal and less scorching during grinding. Atomic force microscopy (AFM), laser scanning digital microscope (LSCM) and SEM results revealed that the ground silicon (Si) wafer had a low surface roughness (~6 nm), without deep scratches, and with a low damaged layer (~0.7 μm) respectively. The wear test on grinding wheels demonstrates it is very cost effective to use a dry processing wheel. This work provides a new method for grinding Si wafers with a new type of wheel being developed by a novel dry superfine diamond grinding technique in a manufacturing process. It is realized that the use of a dry grinding process can save part of chemical polishing process (CMP) and effectively reduces the polishing time.

Iterative method for obtaining nonuniform grinding-induced residual stress distribution of silicon wafers based on global deformation

Grinding as the main thinning process for silicon wafers is widely used in chip manufacturing. The residual stress distribution is desired to be known to optimize the grinding process. In this paper, an iterative method for obtaining the nonuniform residual stress of ground silicon wafers was proposed. The deformation of ground silicon wafers was divided into two parts in which one part was large spherical deformation and the other was a small out-of-sphere deformation. The spherical deformation which was in the nonlinear range corresponded to an average stress which was obtained using the bisection method. The small out-of-sphere deformation corresponded to the stress variation which was obtained using the regularization method with continuity constraints. An iterative successive approximation process was conducted to address the nonlinear problem which was transformed into a local linear problem. The residual stress distribution of a ground silicon wafer was obtained successfully using the proposed method. Although the overall shape of the wafer deformation was approximately spherical, the residual stress value varied a lot with the crystal orientation of the silicon wafer. The stress values in the <110> orientations were much larger than those in the <100> orientations. The method could be used to evaluate the grinding process independently of the dimension sizes of ground silicon wafers. The proposed method in this paper is non-destructive and could be used to track the residual stress changes in the grinding process and the process parameters could be optimized accordingly.

High surface integrity fabrication of silicon wafers using a newly developed nonwoven structured grind-polishing wheel

Silicon wafer is a predominant substrate material in integrated circuits (IC) manufacturing. Currently, grinding is employed as a major machining method for back-thinning and flattening of the wafers. To obtain high surface/subsurface quality of ground silicon wafer, a novel nonwoven structured grind-polishing wheel with ultrafine abrasives is developed in this work. The developed wheel uses zirconia (ZrO2) with a grit diameter of 200 nm as abrasives, where the nonwoven fibers are employed as structures to support the abrasives. By measuring the surface and subsurface quality of ground silicon, grinding performance of the developed grind-polishing wheel is investigated and compared to that of #8000 diamond wheel, which is a commercial grinding wheel with the smallest grit size. The grinding experimental results show that the surface produced by the developed wheel has a surface roughness Ra of 0.45 nm, and a thickness of subsurface damage layer about 67 nm. In addition, the material removal rate for silicon wafers machined by the developed wheel is 0.86 μm/min. Compared with the #8000 diamond wheel, the developed wheel is more suitable for the semi-fine finishing process to produce a smooth surface at reduced time and cost.

Quantifying the subsurface damage and residual stress in ground silicon wafer using laser ultrasonic technology: A Bayesian approach

Dispersion nature of surface acoustic waves (SAWs) propagating along the machined surface contains key information about the elastic properties and stress state of the subsurface damage (SSD) layer. This enables the acoustical characterization of SSD. In this paper, we develop a mechanical properties and residual stress estimation technology for the machined surfaces, especially for ground silicon wafer, based on laser-excited surface acoustic waves (LSAWs). Bayesian inversion is applied to quantify nine orthotropic elastic constants and one residual stress parameter of the SSD layer together with the corresponding uncertainty estimates from dispersion data. The forward model of SAW propagation is derived based on Biot’s theory of small deformations influenced by initial stress. The solution of the inverse problem is given in terms of the posterior probability density (PPD) of the SSD model parameters. Markov chain Monte Carlo (MCMC) method is used to generate posterior samples by repeatedly solving the forward model to compute parameter uncertainties and inter-relationships. Laser ultrasonic experiments have been performed on two typical areas of the ground silicon wafer to determine the ability of the technology to characterize SSD and residual stress and the effect of grinding direction on the damage degree of SSD. The elastic constant values estimated by the proposed method can be reasonably explained by the fracture properties of single-crystal silicon. The reliability of the results is further verified via nanoindentation and micro-Raman spectroscopy.

Sensitivity of polarized laser scattering detection to subsurface damage in ground silicon wafers

Silicon is the primary substrate material in the semiconductor industry. Since surface integrity of a silicon wafer is crucial to the performance of an IC chip made of the wafer, the subsurface damage (SSD) induced by machining to a silicon wafer has been intensively investigated. However, detecting SSD is a challenge due to a lack of an effective method. This study presents a novel method, the polarized laser scattering (PLS) method, for detecting SSD in ground silicon wafers. A PLS system is established to detect the grinding-induced SSD which is also evaluated by the destructive methods. The study shows that the PLS signal is sensitive to the SSD depth with a detection resolution of approximately 0.1 μm. In addition, the detectability of the PLS method is demonstrated and a relationship between the PLS signal and SSD depth is established to facilitate the practical applications of the PLS method.

The influence of subsurface damage depth on fracture strength of ground silicon wafers

A grinding process is required to thin the back of the semiconductor chips to meet the requirements of chip thickness. Subsurface damage caused by the grinding can result in fracture strengths degradation, which has an impact on the safety and reliability of the chip. In this paper, the fracture strengths of silicon wafers ground using wheels with different grit size were obtained by the ball-on-ring method, and the Weibull distribution function was used to fit the fracture strengths. The subsurface damage depths at different radial positions of the ground silicon wafer were measured by cross-section polishing microscopy. Results showed that the fracture strengths decreased with the increase of subsurface damage depths, and the relationship between the fitted minimum strength and the maximum depth of subsurface damage was in good agreement with the prediction result of the classical fracture mechanics formula.

Numerical simulation of laser-generated Rayleigh wave pulses propagation in the machined surface with residual stress using finite-difference method

Laser-excited broadband Rayleigh wave propagation behaviors can be used for characterizing material surface characteristics non-destructively. The present paper is concerned with the Rayleigh wave propagation in the machined surface with a damaged layer with residual stress. Biot’s theory of small deformations affected by initial stress forms the basis for this study. Taking ground silicon wafer as an example, we adopted a staggered finite-difference (FD) scheme for numerical simulation. By comparing the dispersion images with the theoretical dispersion curves of Rayleigh wave, the accuracy of the FD algorithm is verified. Then, some two-layer models without or with residual compressive stress are utilized to further analyze the dispersion energy characteristics of Rayleigh waves in the machined surface. The results indicate that damage layer thickness and damage degree determine the main Rayleigh wave high-order mode and the energy of the high-order mode, respectively. In addition, residual compressive stress can reduce the phase velocity of the Rayleigh wave to a small extent. Meanwhile it can significantly increase the total energy of the Rayleigh wave which excited by the same excitation source. This paper will provide a valuable reference for laser ultrasonic evaluation of residual stress on the machined surface.

Inversion of surface damage and residual stress in ground silicon wafers by laser surface acoustic wave technology

The paper presents a study of surface acoustic waves propagation in a damage layer with finite thickness and residual stress on an orthotropic semi-infinite medium to reveal the application of laser ultrasound in the surface inspection of ground silicon wafers. Biot’s theory of small deformations influenced by initial stress forms the basis for this study. Considering the case that the displacement and boundary forces are continuous at the interface and the forces vanish on the free surface, the required dispersion relation is obtained. We consider a sample of (100) silicon wafer by grinding with fine abrasive grains, which has a machined face with a micrometer-level thickness of surface damage and residual stress. In order to discuss the impact of propagation directions, degree of surface damage, residual compressive stress on the velocity characteristics of surface waves, the numerical computation of the dispersion equations is performed. It has been found that surface damage has a significant effect on the dispersion curve, while the residual compressive stress can only cause a small decrease Δc in the surface wave velocity. The velocity decrease Δc becomes obvious at high frequencies. For a fixed residual compressive stress and frequency, Δc hardly changes with the degree of surface damage and propagation directions. Based on the above characteristics, we study the inverse problem on detecting both surface damage and residual stress simultaneously by SAW velocities and give a corresponding iterative algorithm. This study may provide theoretical guidance for non-destructive testing of residual stress.

Two-dimensional detection of subsurface damage in silicon wafers with polarized laser scattering

Abstract This study proposes a polarized laser scattering (PLS) method for detection of the two-dimensional distribution of the grinding-induced subsurface damage (SSD). This study also uses two destructive methods to observe the distribution of SSD in the ground silicon wafers by cross-sectional microscopy and by layer-by-layer microscopy. An experimental investigation is arranged for detecting the two-dimensional damage from diamond wheel grinding of silicon wafers through wafer sample preparation and SSD observations with the cross-sectional and layer-by-layer microcopy. It also includes PLS detection of the ground silicon wafers for SSD information. The results show that both the depth of SSD and the crack density contribute to the PLS detection signals. The study reveals the characteristics of the PLS detection, which illustrates the potential application of the PLS method to in-process detection.

Subsurface Damage Detection on Ground Silicon Wafers Using Polarized Laser Scattering

Abstract A silicon wafer is important for the electronic and computer industries. However, subsurface damage (SSD), which is detrimental to the performance and lifetime of a silicon chip, is easily induced in a silicon wafer during a grinding process since silicon is typically a hard and brittle material. Therefore, it is necessary to detect and remove SSD in the subsequent processes. In this study, a polarized laser scattering (PLS) system is installed to detect the SSD in a ground wafer. It is found that not only the subsurface crack but also the residual stress leads to depolarization of an incident light. The effects of residual stress on depolarization are studied. The residual stress results in the photoelasticity, which causes the depolarization of the incident light in the PLS system. The depolarization caused by the residual stress is determined by the directions and the difference of the principal stresses. When the polarization direction of the incident light is aligned with one of the principal stresses, the effects of the residual stress can be minimized; therefore, the subsurface crack can be quantitatively estimated by PLS.

Accurate determination of bifurcation points for ground silicon wafers considering anisotropy using FEM method

Silicon wafers warp under residual stress when they are thinned by grinding or plated with a thin layer of a different material. Bifurcation and geometric instability of silicon wafers occur at critical residual stresses that serve as the basis for evaluating the significance of large deformations. Analytical approach based on minimizing the total strain energy is often used to determine the bifurcation points. However, the curvature is assumed uniform across the wafer, which deviates from actual scenarios because of large deformation and anisotropic effects of monocrystalline silicon. In this study, the finite element method considering the anisotropy of silicon was applied to examine the deformation process. Then the surface features of deformed silicon wafer were extracted by employing sphere fitting and removing techniques to the wafer deformation shapes. Based on the variation rule of the deformation features, line fitting technique was applied and the residual stress of the bifurcation point was obtained as the intersection point. The critical value of the stress parameter, which is an important parameter for silicon wafers, was obtained as 650.7 GPa. The bifurcation points of silicon wafers with other dimensions could be obtained easily based on the value with sufficient accuracy, which were verified by FEM results. The proposed method could also be used to determine the bifurcation points of other anisotropic materials.

Generation and distribution of residual stress during nano-grinding of monocrystalline silicon

Residual stress generated in grinding process of monocrystalline silicon can cause the wafer warpage, and difficulties in subsequent processes such as holding and scribing. It can also lead to the formation of cracks and the occurrence of corrosion, which is harmful for electrical performance of silicon component. In this study, with the method of step-wire wet etching, the phase transformation and distribution of residual stress in ground silicon wafer were examined by confocal laser micro-Raman spectroscopy. As the etching depth going down, the residual stress exhibits in the trends of decreasing of compressive stress and following a scatter distribution of tensile stress. During the nano-grinding processes of monocrystalline silicon, the generation mechanism of residual stress is computed by a series of the molecular dynamic (MD) simulation. Subsurface damage (SSD) in the form of phase-transformed silicon is observed, and the depth of SSD varies by the depth of cut. The volume shrinkage of phase-transformed silicon is also studied to explain the grinding mechanism and the reason for inducing residual stress of ground silicon. By adopted the Stony theory and volume shrinkage rate of amorphous phase from MD results, a theoretical model is established to determine the trend of compressive stress in subsurface of ground silicon.

Formation of subsurface cracks in silicon wafers by grinding

Single-crystal silicon is an important material in the semiconductor and optical industries. However, being hard and brittle, a silicon wafer is vulnerable to subsurface cracks (SSCs) during grinding, which is detrimental to the performance and lifetime of a wafer product. Therefore, studying the formation of SSCs is important for optimizing SSC-removal processes and thus improving surface integrity. In this study, a statistical method is used to study the formation of SSCs induced during grinding of silicon wafers. The statistical results show that grinding-induced SSCs are not stochastic but anisotropic in their distributions. Generally, when grinding with coarse abrasive grains, SSCs form along the cleavage planes, primarily the {111} planes. However, when grinding with finer abrasive grains, SSCs tend to form along planes with a fracture-surface energy higher than that of the cleavage planes. These findings provide a guidance for the accurate detection of SSCs in ground silicon wafers.