ArticlePDF Available

Effect of mechanical surface damage on Silicon wafer strength

Authors:

Abstract

Solar power generation using polycrystalline silicon wafers has been rapidly growing in recent years. As a result, it is required to understand the strength characteristics of polycrystalline silicon wafers in order to enhance their quality. Scratches and material defects should be taken into consideration when strength characteristics of polycrystalline silicon are evaluated, since it is a brittle material. In this paper, bending strength of polycrystalline silicon wafers for solar cells were measured, and evaluation regarding the cause of different strength values, which depend on manufacturing conditions of the wafer, was conducted based on fracture mechanics. Residual stress measurements using Raman spectroscopic and observation with TEM (Transmission Electron Microscope) were also conducted. The results clarified the existence of numerous cracks on the wafer surface that are assumed to be generated during slicing process. Thus, it was confirmed that wafer strength depends on the level of machining damage in slicing process. We can establish high reliability for PV modules as a result of modifying the slicing conditions to minimize the mechanical surface damage on wafers and increase the wafer strength.
doi:10.1016/j.proeng.2011.04.239
A
vailable online at www.sciencedirect.com
ICM11
Effect of mechanical surface damage on
Silicon wafer strength
Daisuke Echizenya
ab
*, Hiroo Sakamoto
a
, Katsuhiko Sasaki
b
a
Mitsubishi Electric Corporation, 8-1-1, Tsukaguchi-Honmachi, Amagasaki, Hyogo 661-8661, Japan
b
Hokkaido University, 13-8, kita, Sapporo,Hokkaido 060-8628, Japan
Abstract
Solar power generation using polycrystalline silicon wafers has been rapidly growing in recent years. As a result, it is
required to understand the strength characteristics of polycrystalline silicon wafers in order to enhance their quality.
Scratches and material defects should be taken into consideration when strength characteristics of polycrystalline
silicon are evaluated, since it is a brittle material. In this paper, bending strength of polycrystalline silicon wafers for
solar cells were measured, and evaluation regarding the cause of different strength values, which depend on
manufacturing conditions of the wafer, was conducted based on fracture mechanics. Residual stress measurements
using Raman spectroscopic and observation with TEM (Transmission Electron Microscope) were also conducted.
The results clarified the existence of numerous cracks on the wafer surface that are assumed to be generated during
slicing process. Thus, it was confirmed that wafer strength depends on the level of machining damage in slicing
process. We can establish high reliability for PV modules as a result of modifying the slicing conditions to minimize
the mechanical surface damage on wafers and increase the wafer strength.
Keywords: Crystalline silicon; Fracture strength; Slicing process; Raman spectroscopic; Crack; Residual stress
1. Introduction
In recent years, photovoltaic power generation has been extending all over the world in consideration
to environmental problems. Polycrystalline silicon is widely used since it is cost effective, and this
tendency is expected to continue [1].It is necessary to understand the strength characteristics of silicon
* Corresponding author. Tel.: +81-6-6497-7556
E-mail address: Echizenya.Daisuke@dx.MitsubishiElectric.co.jp.
Procedia Engineering 10 (2011) 1440–1445
1877-7058 © 2011 Published by Elsevier Ltd.
Selection and peer-review under responsibility of ICM11
Open access under CC BY-NC-ND license.
© 2011 Published by Elsevier Ltd.
Selection and peer-review under responsibility of ICM11
Open access under CC BY-NC-ND license.
Daisuke Echizenya et al. / Procedia Engineering 10 (2011) 1440–1445 1441
wafers in order to enhance the quality (ex. the thinner the wafer, reliability for a longer lifetime) of PV
modules.
Crystalline silicon is a brittle material and has high notch sensitivity. Therefore, crack evaluation is
indispensable in its quantitative strength assessment, and especially attention should be paid to
manufacturing damage in the slicing process from ingot to wafer production. In the past, there were some
reports of strength evaluation of silicon wafer for integrated circuits [2]. Because the manufacturing
process of wafers for solar cells is not the same as for integrated circuits it needed a single purpose
evaluation for polycrystalline silicon. Especially polycrystalline silicon has many crystal grain boundaries
and deficiencies in the crystal structure. It means that the clarification of a strength factor is important for
the polycrystalline silicon.
This paper presents the effects of surface damage on the silicon wafer strength. The polycrystalline
silicon wafers were manufactured by a wire saw. It is known that the affected layer of the surface occurs
when the ingot is sliced (Fig 1). First we measured strength and fracture toughness, so that the strength of
damaged polycrystalline silicon wafer was evaluated based on fracture mechanics assuming the damaged
layer as potential cracks on the wafer surface. For evaluation of surface damage, microscopic observation
by using both a transmission electron microscope (TEM) and Raman spectroscopic were conducted for
different strength polycrystalline silicon wafers.
2. Evaluation of
5
trength Characteristics
2.1. Test Samples
A polycrystalline silicon wafer for this study is shown in Fig 2. The sample was manufactured by
slicing an ingot into 150mm x 150mm square wafers of 200ȝm to 300ȝm thickness using a wire saw, and
then cutting 10mm widths using diamond dicing. The cut surface by diamond dicing is shown in Fig 3.
The cut surface is smoother compared with the wafer surface and it was confirmed that fracture did not
initiate from the cut surface. Three types of test pieces were prepared: samples are manufactured by
different slicing conditions. Sample A and B were made from the same ingot, but sample C was made
from the different ingot.
2.2. Bending Strength Test
A four point bending test was conducted to measure the tensile strength by using a tensile testing
machine (Fig 4). A four point bending test can get more accurate results than a three point bending test
and ring-on-ring test. Because it can apply high stress over a large area, there is little displacement from
the high stress point to the break point. Bending strength ı was calculated using the following equation,
2
)(3 21
2
LL
bt
F
V
(1)
Where, F is the applied load, b is the width of the sample, t is the thickness, L1 is the larger span, L2 is
the smaller span.
1442 Daisuke Echizenya et al. / Procedia Engineering 10 (2011) 1440–1445
Fig 5 shows measured results of bending strength. Samples A, B, and C have different bending
strengths. Especially, the bending strength of samples A and C are different value although samples A
and C are made from the same ingot. It suggests that the difference in bending strength is not caused by
the difference in ingots but caused by slicing damage on the surface.
2.3. Fracture Toughness Test
In order to evaluate the effect of surface cracks due to machining damage, fracture toughness (KIC) of
polycrystalline silicon wafers was measured by controlled surface flaws (CSF) method [3]. Same test
piece for bending strength test was used. Knoop indenter was pressed on the center of the test piece with
its longitudinal axis set to be perpendicular to the longitudinal direction of the test piece. Indentation load
was set at 5N and 10N. Sample A and sample B from different ingots were used. Then, the four point
bending test was conducted by loading a tensile stress on the surface where the Knoop indentation had
been applied. KIC was calculated by following equation for stress intensity factor,
KIC
˙
F
V
Ba
S
(2)
where ıB is the fracture stress obtained from the bending strength test, a is the crack depth due to Knoop
indentation, and F is a correction factor, respectively. Semi-elliptic crack due to Knoop indentation can be
seen and F is determined from the following equation [3].
F=1.1359-0.3929ȝ-0.3440ȝ2-0.2613ȝ3+Ȝ(-1.5184+0.4178ȝ+0.7846ȝ2-0.6329ȝ3)
+Ȝ2(4.3721-13.9152ȝ+16.2550ȝ2-6.4894ȝ3)+Ȝ3(-3.9502+12.5334ȝ-14.6137ȝ2+5.8110ȝ3)
and
ȝ=d/w, Ȝ=d/t (3)
Slurry
Affected
layer
Wire
Wire
150mm
Fig. 1. Image of affected layer when wire slicing Fig. 2. silicon wafer Fig. 3. Test piece
L1=30mm
L2=10mm
Fig. 4. Four-point bending test
Daisuke Echizenya et al. / Procedia Engineering 10 (2011) 1440–1445 1443
Where, d is crack depth, w is a half crack width, and t is thickness of the test piece.
Although silicon is an anisotropic material, measured KIC for different crystal orientations are
distributed from 1.11±0.07 to 1.18±0.03 MPa m0.5 according to the previous investigation [3], which is
little affected by crystal orientation. Therefore, crystal orientation was not taken into consideration in this
investigation.
Fig 6 shows measured results of fracture toughness (KIC). Fig 7 shows the fracture surface. Sample A
and B have almost same KIC values. Though the bending strength of sample A and B is different they
have almost same the KIC value. Therefore, it can be considered that the difference in bending strength is
caused by slicing damage on the surface as well.
40 60 80100 300
.01
.1
1
5
10
20
30
50
70
80
90
95
99
99.9
99.99
0.2 0.4 0.6 0.8 1
.01
.1
1
5
10
20
30
50
70
80
90
95
99
99.9
99.99
Fig. 5. Results of strength measurement Fig. 6. Results of fracture toughness measurement Fig. 7. Controlled flaw
2.4. Strength Test for the Wafer Eliminated Damage
We manufactured a sample whose surface damage was eliminated then we conducted a bending
strength test described in chapter 2.2. Sample D was manufactured from sample A by surface polishing
and reducing 20% of its thickness. Fig 8 shows the polished sample compared with an unpolished sample.
The results are shown in Fig 9. The bending strength of the sample D is about three times greater than
that of sample A in the initial condition, confirming the big contribution of surface cracks to the strength
deterioration of polycrystalline silicon wafers
Therefore this result shows that the crystal grain boundary and deficiency of the crystal structure of
polycrystalline silicon does not affect the strength of the wafer at least under 300MPa.
 10 100 1000
.01
.1
1
5
10
20
30
50
70
80
90
95
99
99.9
99.99
Fig. 8. Polished sample Fig. 9. Results of strength measurement
Depth˖a
Width˖2b
10Pm
Initial crack by indentation
Cumulative probability(%)
Fracture toughness(MPa m0.5)
Cumulative probability(%)
Strength(MPa)
Sample A
Sample B
Sample C
Sample D
Cumulative probability(%)
Strength(MPa)
sample A(113MPa)
sample B(122MPa)
sample C(89.2MPa)
sample A(0.56MPa m0.5)
sam
p
le B
(
0.58MPa
m
0.5
)
Bare
Polished
1444 Daisuke Echizenya et al. / Procedia Engineering 10 (2011) 1440–1445
3. Evaluation of Surface Damage
3.1. Surface Observation by TEM
Surface observation by TEM was conducted for sample B and C in order to investigate the level of
surface damage. The sample for the observation was cut by Focused Ion Beam (FIB) to observe the
thickness direction of the wafers. Average sample thickness was 0.25 ȝm, and length was 20 ȝm.
Results of TEM observation is shown in Fig 10 and Fig 11 at the same scale. In spite of a narrow
observation surface of a couple of micron square meters, some interference fringes that indicated crack
existence were observed in both samples. Furthermore, the crack in sample C is apparently larger than
that of sample B, which indicates a correlation with bending strength.
 
Fig. 10. TEM image of cross section (sample B) Fig. 11. TEM image of cross section (sample C)
3.2. Residual Stress Measurement using a Raman Spectroscopic
Observed area by TEM is 10-8 times smaller compared to test piece area of strength sample. Therefore,
evaluation of residual stress distribution using Raman spectroscopic was tried in order to cover a wide
surface range. Raman spectroscopic analysis can measure residual stress as mechanical damage, and
evaluate the deterioration of the crystal structure [5]. Total value of three axis principal stress is measured
as a positive value of compressive stress in the Raman spectroscopic. Residual stress from the surface to
the thickness direction was measured in the wide cross section.
Measured residual stress for sample A and D are shown in Fig 12. In Fig. 12 axis of ordinate shows
residual stress and abscissa is a distance from the surface normalized by sample A thickness. The
compressive residual stress near the surface of sample A is high, while that of sample D is low due to
mirror polishing. Therefore, it was confirmed that damage by machining causes the residual stress.
Bending strength of sample D is higher while its compressive residual stress near the surface is low. It
can be considered that the singular stress field caused by the crack chapter has a greater effect on bending
strength than the enhanced strength caused by the compressive residual stress.
The residual stress of samples A, B, and C are shown in Fig 13 as a log-log plot. Residual stress
converges to some extent as the distance from the surface becomes greater. It was confirmed from
samples A, B, and C that surface machining damage has an effect on the wafer bending strength because
the bending strength reduces as the residual stress increases.
Focusing on the residual stress near the surface of samples A, B, and C described in Fig 13, it can be
assumed that transitions or cracks exist due to machining damage and that crystal structure is deteriorated
because the residual stress has a large variation. The maximum residual stress had a tendency to saturate
near 500 MPa for all samples, clarifying the existence of a residual stress threshold. This phenomenon is
probably due to the nonlinearity of crystal silicon such as the occurrence of transitions caused by shear
Wafer surface
Crack
Crack
d=0.01
Wafer surface
Thickness
direction
Crack
d=0.01
Daisuke Echizenya et al. / Procedia Engineering 10 (2011) 1440–1445 1445
stresses or the occurrence of cracks caused by the destruction of crystals. These transitions or cracks are
supposed to work as potential cracks [6] but work as strength parameters instead.
-100
0
100
200
300
400
500
00.02 0.0 4 0.06 0.08
0.1
1
10
100
1000
0.001 0.01 0.1
Fig. 12. Results of Raman spectroscopic (Linear scale) Fig. 13. Results of Raman spectroscopic (Log scale)
4. Conclusions
Investigations on surface damage and strength measurement of photovoltaic polycrystalline silicon
wafers were conducted and the following results were obtained:
(1) According to the bending strength test and the fracture toughness test, surface damage due to
machining was found to affect bending strength. The bending strength of the samples whose machining
damage was modified by mirror polishing was about three times higher than that of samples in the initial
condition. Also it was found that the crystal grain boundary and deficiency of the crystal structure of
polycrystalline silicon does not have an effect on the strength of the wafer at least under 300MPa.
(2) Surface observation by TEM clarified numerous cracks on wafer surface. It also showed that cracks
on a small strength sample have a strength that is apparently larger than that of a large strength sample,
which indicates correlation with bending strength.
(3) From surface observation by Raman spectroscopic it was confirmed that damage by machining
caused the residual stress.
The results clarified the existence of numerous cracks on the wafer surface that are assumed to be
generated during the slicing process. Thus, it was confirmed that wafer strength depends on the level of
machining damage in the slicing process. We can predict higher reliability for PV modules as a result of
modifying the slicing conditions so as to reduce the mechanical surface damage on wafers and thereby
increase the wafer strength.
References
[1] Photovoltaic Power Generation Market; Present Trend of Enterprises, Journal of Architectural Products, 2004, 283,p.21-24.
[2] S. Takyu, et al., Novel Wafer Dicing and Chip Thinning Technologies Realizing High Chip Strength, Electronic
Components and Technology Conference, 2006; p1623-1627.
[3] K. Hayashi, et al., Fracture toughness of single crystal silicon, Journal of the Society of Materials Science, Japan, vol. 40,
No.451,1991, p. 39-44
[4] The Society of Materials Science, Japan, STRESS INTENSITY FACTORS HONDBOOK Volume 2, 1987,p. 698-699
[5] E.Anastassakis, Effect of Static Uniaxial Stress on the Raman Spectrum of Silicon, Solid State Com., vol.8, 1970, p.133-138
[6] K. Tanaka, Y. Nakai, Fatigue growth threshold of small cracks, Int. Journ. of Fractureˈvol.17, No.5, 1981,p.519-533 .
Ccmpressive residual stress(Mpa)
Ccmpressive residual stress(Mpa)
Normalized distance from surface
sample A(113MPa)
sample B(122MPa)
sample C(89.2MPa)
Normalized distance from surface
sample A(Bare)
sample D(Polished)
... There is currently no single criterion for choosing a method for determining the strength of thin silicon wafers. Most researchers take the critical maximum tensile stress during bending, uniaxial (threeor four-point uniaxial bending of strips [1][2][3][4][5][6][7][8]) or biaxial (biaxial) bending of plates as the strength value [9][10][11][12][13][14]. ...
... The surface condition of thin silicon wafers is the most important factor determining their strength. The impact of the conditions for obtaining and processing thin silicon wafers on their strength was addressed in the papers listed above [1][2][3][4][5][6][7][8][9][10][11][12][13][14][15][16]. Any change in these conditions, even such as the place in the plate in the ingot from where it was cut, the alloying of the plate material with controlled or background impurities, the location of a small chip on an initially large plate affect the strength of the sample. ...
Article
Full-text available
The mechanical strength of various silicon wafers with a thickness of 100 μm has been studied, depending on the methods of their preparation and the modes of their subsequent grinding or polishing, including chemical-mechanical (HMP). The plates were loaded using the ring-to-ring method, the magnitude of stresses and deflection under the small ring was determined by the finite element method. For all the samples studied, the profiles and roughness parameters of the plates were obtained by stylus profilometry and atomic force microscopy (AFM) when scanning the surface along the baseline and over the area. A direct correlation was found between the strength of the plates and the characteristic parameters of their surface profile (the average values of the magnitude and period of fluctuations in the height of the irregularities). Keywords: silicon wafer, strength, chemical-mechanical polishing treatment, surface morfology.
... Единого критерия для выбора способа определения прочности тонких пластин кремния в настоящее время нет. Большинство исследователей за величину прочности принимают критическое максимальное растягивающее напряжение при изгибе, одноосном (трехили четырехточечный одноосный изгиб полосок [1][2][3][4][5][6][7][8]) или двухосном (биаксиальном) изгибе пластин [9][10][11][12][13][14]. Значительно реже используются другие виды воздействия, например, индентирование [15] и разрушение в результате термического расширения при нагревании, стимулирующего развитие трещин [16]. Аналитические выражения для максимальных напряжений при изгибе были заимствованы из работ по разрушению других хрупких тел (стекла, керамики и др.). ...
... 1. Состояние поверхности тонких пластин кремния является важнейшим фактором, определяющим их прочность. В перечисленных выше работах [1][2][3][4][5][6][7][8][9][10][11][12][13][14][15][16] рассматривалось влияние условий получения и обработки тонких пластин кремния на их прочность. Любое изменение этих условий, даже таких как место в пластины в слитке, откуда она была вырезана, легирование материала пластины контролируемыми или фоновыми примесями, место расположения чипа малого размера на исходно большой пластине сказываются на величине прочности образца. ...
Article
Full-text available
The mechanical strength of various silicon wafers with a thickness of 100 µm has been studied, depending on the methods of their preparation and the modes of their subsequent grinding or polishing, including chemical-mechanical (HMP). The plates were loaded using the ring-to-ring method, the magnitude of stresses and deflection under the small ring was determined by the finite element method. For all the samples studied, the profiles and roughness parameters of the plates were obtained by stylus profilometry and atomic force microscopy (AFM) when scanning the surface along the baseline and over the area. A direct correlation was found between the strength of the plates and the characteristic parameters of their surface profile (the average values of the magnitude and period of fluctuations in the height of the irregularities).
... The Analysis of finite element analysis of a model of the same dimensions and material properties utilizing 8-noded bi-quadratic plane strain element (CPS8R) and J-integral also agrees with the outcome obtained through discrete molecular simulations. Furthermore, the results presented here demonstrate concurrence with reported outcomes obtained through critical energy release rate utilizing molecular dynamics (MD) calculations [31], and experiments done in thin films of pure single silicon [32] and poly-crystalline crystals [33]. The consistency of outcomes shown in Table 2 affirms the validity of the findings in this study, reinforcing the alignment between theoretical predictions made in this work and experimental data. ...
Conference Paper
Full-text available
The investigation of material behavior at the atomic scale, especially concerning crack propagation, has presented enduring challenges in the fields of materials science and engineering mechanics. The current progress in nanotechnology has amplified the demand for precise characterization of stress and strain fields in the vicinity of crack tips. Such comprehension is vital for deciphering intricate material behavior during crack propagation and fracture mechanisms, ultimately contributing to advancements in materials design and the augmentation of mechanical properties. This research addresses this demand by introducing a deformation mapping approach that provides unique insights into strain field characterization for both crystalline and amorphous multi-atomic materials. While molecular statics have been used to obtain stress fields in various materials, accuracy diminishes when dealing with amorphous materials, internal defects, or complex structures. Silicon, a material with wide-ranging applications, serves as a suitable subject for this study, aiming to bridge the gap between theory and experimentation. This work focuses on the applicability of linear and quadratic mapping approaches, with varying homogenization zone radii. Additionally, it aims to estimate the Mode-I fracture parameter KIC and compare the results with finite element analysis, Voronoi volume analysis, and other experimental and analytical findings. This work represents a significant step towards advancing our understanding of crack growth at the molecular scale and its applications in materials engineering.
... They also found that the phase distribution during the reverse transformation was strongly influenced by the indenter radius and indentation depth due to the different stress fields. Echizenya et al. [10] established a function relationship between stress and etching depth, and the results showed that the residual compressive stress decreased with the increase of wafer depth. Wurzner et al. [11] used one single abrasive to scratch monocrystalline silicon and detected the residual stress through Raman spectroscopy. ...
Article
Full-text available
Monocrystalline silicon wafer is the most important raw material in chip manufacturing; wire sawing is the most common processing method of monocrystalline silicon wafer. The residual stress generated by cutting affects the subsequent polishing costs directly, as well as the fracture strength and mechanical integrity of the monocrystalline silicon wafer, thus affecting the service performance. In this paper, the generation mechanism of residual stress was analyzed based on diamond wire sawing technology. Then, based on Drucker-Prager (D-P) plastic constitutive model, the magnitude and distribution of residual stress of monocrystalline silicon chip under different process parameters were simulated by ABAQUS simulation software. Finally, the experiment of ultrasonic vibration–assisted diamond wire saw cutting monocrystalline silicon was conducted, and the residual stress of monocrystalline silicon chips were detected by blind hole method. During the detection, the strain values before and after drilling in three directions of monocrystalline silicon chips were measured by strain gauge rosette, and then the residual stress values were determined according to the strain values and the residual stress calculation theory. The experimental results showed that residual compressive stress remains on the surface of silicon wafer in both conventional wire sawing and ultrasonic vibration–assisted wire sawing. The residual compressive stress increases with the increase of the axial speed of wire saw, decreases with the increase of the wire saw feed speed, and fluctuates with the increase of the workpiece rotation speed. The residual compressive stress on the surface of the silicon wafer by ultrasonic vibration–assisted wire sawing is larger than that by conventional wire sawing. The simulation and experimental results have the same trend, which verifies the validity of the finite element model in this paper and provides theoretical support for the subsequent processing of single crystal silicon.
... They also found that the phase distribution during the reverse transformation was strongly influenced by the indenter radius and in-dentation depth due to the different stress fields. Echizenya et al. [10] established a function relationship between stress and etching depth, and the results showed that the residual compressive stress decreased with the increase of wafer depth. Wurzner et al. [11] used one single abrasive to scratch monocrystalline silicon and detected the residual stress through Raman spectroscopy. ...
Preprint
Full-text available
Monocrystalline silicon wafer is the most important raw material in chip manufacturing, wire sawing is the most common processing method of monocrystalline silicon wafer. The residual stress generated by cutting affects the subsequent polishing costs directly, as well as the fracture strength and mechanical integrity of the monocrystalline silicon wafer, thus affecting the service performance. In this paper, the generation mechanism of residual stress was analyzed based on diamond wire sawing technology. Then, based on D-P plastic constitutive model, the magnitude and distribution of residual stress of monocrystalline silicon chip under different process parameters were simulated by ABAQUS simulation software. Finally, the experiment of ultrasonic vibration assisted diamond wire sawing monocrystalline silicon and blind hole drilling method detection were conducted, the strain values before and after drilling in three directions were measured by strain gauge rosette, and the residual stress was calculated by combining with mathematical theory, and the experimental results verified the validity of the finite element model. The experimental results showed that residual compressive stress remains on the surface of silicon wafer in both conventional wire sawing and ultrasonic vibration assisted wire sawing. The residual compressive stress increases with the increase of the axial speed of wire saw, decreases with the increase of the wire saw feed speed, and fluctuates with the increase of the workpiece rotation speed. The residual compressive stress on the surface of the silicon wafer by ultrasonic vibration assisted wire sawing is larger than that by conventional wire sawing. The variation trend of simulation and experiment results is consistent, which provides theoretical support for the subsequent processing of monocrystalline silicon wafers.
Article
Full-text available
Silicon (Si) has been widely investigated as a feasible material for photoelectrochemical (PEC) water splitting. Compared to thick wafer-based Si, thin Si (< 50 μm thickness) could concurrently minimize the material usage allowing the development of cost-effective and flexible photoelectrodes for integrable PEC cells. This work presents the design and fabrication of thin Si using crack-assisted layer exfoliation method through detailed optical simulations and a systematic investigation of the exfoliation method. Thin free-standing Si photoanodes with sub-50 μm thickness are demonstrated by incorporating a nickel oxide (NiOx) thin film as oxygen evolution catalyst, light-trapping surface structure, and a rear- pn⁺ junction, to generate a photo-current density of 23.43 mA/cm² with an onset potential of 1.2 V (vs. RHE). Our work offers a general approach for the development of efficient and cost-effective photoelectrodes with Si films with important implications for flexible and wearable Si-based photovoltaics and (opto)electronic devices.
Article
Ultra-thin chips (UTCs) are needed to meet the performance and packaging related requirements of flexible electronics and 3D integrated circuits (ICs). However, handling of UTCs (<50 μm thick), particularly after thinning, is a challenging task as the excessive mechanical stresses could lead to cracking. Such damages could be prevented by restricting the stresses to acceptable levels. Herein, we present a new reliable and cost-effective method based on a polymethylmethacrylate (PMMA) sacrificial layer (20 μm-thick). The PMMA layer results in 4 order of magnitude lower stress on UTCs and, as a result, the reliable removal or debonding of UTCs (35 μm-thick) from the glass substrate has been achieved. The distinctive features of the presented method are high reliability and cost-effectiveness (an order of magnitude cheaper) with respect to conventional methods that use UV curable tapes. The UTCs with metal-oxide-semiconductor capacitors (MOSCAPs) devices were also obtained using this approach and were evaluated under different bending conditions. The stable and uniform performance (134 pF) observed under bending conditions demonstrates that the presented technique could be useful for integration of high-performance flexible UTCs on flexible printed circuit boards for various practical application.
Article
As telecommunication equipment that supports high-level information networks is being made portable, the requirements for telecommunication equipment to be small and lightweight are becoming stricter. Thus, miniaturization of semiconductor devices is necessary, and wafer dicing and chip thinning technologies are important key technologies to achieve it. Wafers are thinned by mechanical in-feed grinding using a grindstone containing diamond particles, and wafers are divided by mechanical blade dicing using a diamond blade. However, mechanical processes using diamond grits leave damage such as chipping, saw mark or residual strain on chip surfaces; thus, chip strength decreases. At chip thicknesses of 50 to 200 μm, such damage has to be avoided. In this study, the relationship between chip residual damage and chip strength is examined, and novel wafer dicing and thinning technologies that realize an average chip strength have increased from 253 MPa to 1903 MPa are described.
Article
Fracture toughness (KIC) values were measured on the major planes of single-crystal silicon using the controlled surface flaw (CSF) and identation fracture (IF) methods at room temperature. The values of KIC on the surface orientations {100}, {110} {111} and {112} obtained by the CSF method were found to be 1.13, 1.12 approximately 1.14, 1.18 and 1.11 approximately 1.18 MPa · m0.5, respectively. It was found that these KIC values increased slightly with increasing the Young's modulus perpendicular to the fracture surfaces. Though KIC values obtained by the IF method, on the other hand, were scattered significantly (0.82 approximately 1.18 MPa · m0.5), its Young's modulus dependence was similar to that obtained by CSF method.
Article
We have performed the first experimental investigation of the effects of static uniaxial stress on the frequency of the q ≈ 0 optical phonons in a diamond-type crystal. Application of uniaxial stress along [001] or [111] is found to cause splittings and shifts of the Raman peaks of the q ≈ 0 triply degenerate optical phonons. From the observed splittings we have obtained experimental values for the phenomenological coefficients which describe the changes in the ‘spring constant’ of these phonons with strain. Our results are in reasonable agreement with lattice dynamical theory.
Article
We have performed the first experimental investigation of the effects of static uniaxial stress on the frequency of the q≈0 optical phonons in a diamond-type crystal. Application of uniaxial stress along [001] or [111] is found to cause splittings and shifts of the Raman peaks of the q≈0 triply degenerate optical phonons. From the observed splittings we have obtained experimental values for the phenomenological coefficients which describe the changes in the ‘spring constant’ of these phonons with strain. Our results are in reasonable agreement with lattice dynamical theory.RésuméNous avons effectué les premières recherches expérimentales des effets de tension uniaxiales statiques sur la fréquence des phonons optiques avec q≈0 dans un crystal du type du diamant. Nous avons trouvé que l'application de tension uniaxiale dans les direction [001] ou [111] cause des divisions et deplacements des bande de l'effet Raman des phonons optiques triplement dégénérés avec q≈0. Des division observées nous avons calculé les valeurs expérimentales pour les coéfficients phénoménologiques qui decrivent les changements des constantes de forces de ces phonons avec le déformation. Nos résultats confirment raisonnablement la theorie dynamique des réseaux.
Article
The effects of grain size and crack length on the threshold condition of fatigue growth of small cracks are analysed both theoretically and experimentally. The theoretical model is based on the assumption that the threshold condition is determined by whether the crack-tip slip band blocked by the grain boundary propagates into an adjacent grain or not. By comparing the model analysis with the experiments of mild steel, it was found that the threshold stress estimated from the fatigue limit of the smooth specimen gave the effective components of the applied stress. When the threshold stress normalized by the fatigue limit of the smooth specimen was correlated to the crack length normalized by a certain crack length, the theoretical relation was found to agree well with the experimental data. From a further analysis of the published data of various materials, the theoretical normalized relation was found to give a lower bound of the threshold values of the stress and the stress intensity factor. A deviation from the theoretical relation seen in the cases of hard metals was explained through an extension of the model by regarding that hard metals contained initial, inherent flaws even in the smooth specimen.
Conference Paper
As telecommunication equipment that supports high-level information networks is being made portable, the requirements for telecommunication equipment to be small and lightweight are becoming stricter. Thus, miniaturization of semiconductor devices is necessary, and wafer dicing and chip thinning technologies are important key technologies to achieve it. Wafers are thinned by mechanical in-feed grinding using a grindstone containing diamond particles, and wafers are divided by mechanical blade dicing using a diamond blade. However, mechanical processes using diamond grits leave damage such as chipping, saw mark or residual strain on chip surfaces; thus, chip strength decreases. At chip thickness of 50 to 200 mum, such damage has to be avoided. In this study, the relationship between chip residue damage and chip strength is examined, and novel wafer dicing and thinning technologies that realize an average chip strength have increased from 253 MPa to 1312 MPa are described
Photovoltaic Power Generation Market; Present Trend of Enterprises
Photovoltaic Power Generation Market; Present Trend of Enterprises, Journal of Architectural Products, 2004, 283,p.21-24.