Open AccessArticle

Investigation of Ultra-Thin Glass Scribing Mechanism

Dawei Li

¹,

Jiahao Li

²,

Huaye Kong

²,

Jinzhu Guo

¹,

Liyong Huang

¹ and

Yao Liu

^1,2,*

CETC Fenghua Information Equipment Co., Ltd., Taiyuan 030024, China

Shanxi Key Lab of Advanced Manufacturing Technology, North University of China, Taiyuan 030051, China

Author to whom correspondence should be addressed.

Coatings 2025, 15(3), 275; https://doi.org/10.3390/coatings15030275

Submission received: 3 February 2025 / Revised: 21 February 2025 / Accepted: 24 February 2025 / Published: 26 February 2025

Download

Browse Figures

Figure 1
Schematic of scribing force. "> Figure 2
Fracture modes. "> Figure 3
Experimental setup. "> Figure 4
Schematic diagram of scribing and breaking process. (a) Glass scribing and breaking process, (b) scribing line, and (c) cross-section view. "> Figure 5
Effect of different scribing wheel angles and scribing forces on the surface topography of glass substrate. (a) θ = 90° and F = 20 N, (b) θ = 120° and F = 20 N, (c) θ = 140° and F = 30 N. "> Figure 6
Cutting cross-section morphology of glass by using (a) θ = 90° and F = 3 N, (b) θ = 90° and F = 10 N, (c) θ = 110° and F = 10 N, (d) θ = 120° and F = 20 N, (e) θ = 140° and F = 20 N, (f) θ = 140° and F = 30 N. "> Figure 7
Effect of scribing force and scribing wheel angle on lateral crack size. "> Figure 8
Effect of scribing force on the propagation rate of lateral cracks. "> Figure 9
Effect of scribing force and scribing wheel angle on cutting micro-crack depth and median crack length (line graph—median crack; bar graph—micro-crack depth). "> Figure 10
Stress variation at the crack tip with different (a) scribing wheel angle and (b) scrining force. "> Figure 11
Height cloud map and 3D views of glass cross-sections at different cutting parameters. (a) surface profile for deflection angle measurement, (b) deflection angle at θ = 130° and F = 16 N, and (c) deflection angle at θ = 90° and F = 10 N. "> Figure 12
(a) Effect of scribing force and scribing wheel angle on cross-section deflection angle at 90–110°; (b) 120°, (c) 130°, (d) 140°; effect of scribing force on median crack and cross-section deflection angle. "> Figure 13
Effect of scribing speed on lateral cracks, median cracks, micro-crack depths, and cross-section deflection angle. (a) lateral crack and deflection angle, (b) microcrack depth and median crack. ">

Versions Notes

Abstract

To reveal the scribing mechanism of ultra-thin glass, single-factor scribing tests were carried out. The effects of the scribing wheel angle θ, scribing force F, and scribing speed v on the lateral cracks width w, scribing depth d, median cracks size l, and cross-section deflection angle α were analyzed to present the scribing quality. The results show that w increases with an increase in θ and F. Further, l and d increase with an increase in F. However, d shows an increasing trend with the increase in θ, and l shows a decreasing trend. In the range of 120–140°, α shows a trend of increasing first and then decreasing with an increase in F. The 120° scribing wheel angle, 20 N scribing force, and 100–400 mm/s scribing speed show the best scribing quality, which limits micro-cracks at the initiation stage without any damage or chipping. Under this condition, the breaking surface edges were free of debris and cracks. A smooth and trim Wallner ripple was obtained from the median cracks with a minimum deflection angle.

Keywords:

scribing wheel; glass scribing; lateral cracks; longitudinal cracks; cross-section quality

1. Introduction

Ultra-thin glass (UTG) is widely used as a substrate for display panels, such as TV, computer, and cellphone screens. The high-generation display panels often have a very large size, which requires diamond scribing and breaking into the suitable product size [1]. The diamond scribing wheel adds a normal force to the glass, which causes a lateral and median crack on the surface. Then, the breaking bar forces the median crack propagation to separate the glass into cells. At the same time, defects, such as crack chipping, debris, uneven lines, burst edges, burr, and scratches, are generated on the surface and scribe line, which may cause glass cell fracture in the following process [2]. Therefore, scribing process optimization to reduce defects is essential.

After a long period of practical exploration, researchers at home and abroad have achieved an in-depth understanding of the scribing wheel selection, force control, and yield rate in glass scribing. Lee et al. [3] proposed that due to the brittle nature of glass, micro-cracks and chips are generated at or near the glass edge or surface, resulting in sensitive glass that is fragile to temperature and mechanical loads due to the cracks, which limits the bending strength and mechanical integrity [4]. An et al. [5] investigated the cutting characteristics of BK7 optical glass and fused silica glass by using nanoscale cutting experiments, and brittle fractures dominated the chip removal mode. Cheng et al. [6], using finite element analysis, found that, mechanically scribed ultrathin glass showed significant edge fragmentation and damage, leading to a significant reduction in bending strength. Murata [7] experimentally investigated the optimized scribing parameters for AMLCD glass. Li et al. [8] proposed a solution for sapphire glass scribing in automated production by analyzing the process and factors that influence its quantity. Luo et al. [9] investigated the characteristics and wear modes of thin diamond scribing wheels in the cutting and grinding of BK7 optical glass and found that lower transverse speeds can obtain better cutting straightness. Li et al. [10] classified the influencing factors that produce defects based on the cutting/breaking principle and carried out an experiment for validation, which improved the cutting quality and yield rate. Zhou et al. [11] identified the factors that lead to variation in the cut size of glass substrates based on the essence of the cut size accuracy control. Pei et al. [12] explored the effect of different parameters on the longitudinal cut marks in cutting, and performed a comprehensive consideration of three parameter—the scribing wheel angle, knife depth, and pressure—to obtain better cutting results. Zhou et al. [13] compared three common scribing wheels and proposed two optimization measures, precision adjustment of scribing wheel pressure and step-by-step breaking, to improve the cutting quality. Ono et al. [14] found that the angle and diameter of the scribing wheel had a significant effect on the median crack depth and proposed an equation to predict the median crack depth. Li et al. [15] discussed the in-line cutting process of ultra-thin glass, emphasizing the influence of key parameters such as the tooth shape of the scribing wheel, the number of teeth, the angle, and the pressure of the scribing wheel on the cutting quality. Chen et al. [16] analyzed the stress distribution under different machining processes, explored the crack extension and rolling brittle fracture quality of different materials by assessing the scribing wheel geometry and process parameters, and found that diamond scribing wheels with a micro-serrated structure have the best cutting quality. Wu et al. [17] used the same scribing wheel-assisted lobing process to reduce the wavy grain defective rate from 0.1% to 0.01% during the cutting of glass with a multi-fluted scribing wheel. Swain et al. [18] resorted to indentation fracture mechanics to determine the effect of load and tool geometry on the median crack depth during glass cutting. Pan et al. [19] investigated the effect of different scribing forces and scribing wheel depths on median crack and lateral crack lengths and found that increasing the scribing force leads to an increase in median crack and lateral crack lengths. Zhang et al. [20] aimed at exploring the phenomenon of glass sharp corner breakage during the intersecting line cutting process by adopting the method of skipping knife cutting at the intercross-section point; breakage was effectively prevented and the production yield was improved. Li et al. [21] provided a detailed description of the causes, countermeasure methods, and countermeasure results of various types of glass adhesion-related defects in the cutting process. Wu et al. [22] explored a liquid crystal glass cutting process dummy (residual material) strip for residual problems, and they achieved the establishment of a dummy monitoring system and interception program, effectively reducing the occurrence of scratches and other adverse effects. Liu et al. [23] optimized process conditions such as the scribing wheel angle, scribing force, and cut-in volume, which resulted in a reduction in glass chips generated on the surface of the glass substrate after cutting/breaking, and a decrease of 1% to 2% in the glass chip foreign matter in the subsequent polarizer attachment process.

Existing studies have focused on the optimization of a few parameters in the macroscopic crack propagation mechanism or the improvement of cutting accuracy based on existing cutting production systems. However, studies on the characteristics of cut quality by analyzing the fracture microstructure of glass cracks during the actual cutting process have not been reported. Based on Griffith’s micro-crack theory and glass fracture crack stress distribution equation, this study aims to carry out a glass substrate cutting experiment; to study the impact of scribing wheel angle, scribing force, and scribing speed on the glass micro-crack expansion and cutting quality; to explore the causes of defects such as chips and cracks during cutting, and to provide a set of reference for the glass substrate cutting process.

2. Mechanical Cutting Principle

In the cutting process, the material hardness of the scribing wheel is greater than the hardness of the glass substrate. When cutting, the scribing wheel makes a continuous scratch, i.e., cutting line, on the glass surface at a fixed scribing force and speed. The purpose is to produce certain micro-cracks on the surface of the glass; as shown in Figure 1, the generation of micro-cracks is related to the size of the angle of the scribing wheel and the magnitude of the applied pressure, and the force on the glass during the cutting is given in Equation (1):

F_{1} = F / 2 \sin θ / 2

(1)

F₁ is the component force perpendicular to the side of the scribing wheel’s cutting edge, Cutting-Edge Side Force (CESF), F is the scribing force perpendicular to the glass surface, and θ is the angle of the scribing wheel.

The end of the micro-cracks is where the stress is concentrated, according to Griffith’s theory of micro-cracks [24]. Under the action of scribing force, there exists a critical crack length of the micro-cracks in the material, and when it exceeds the critical size, the extension force of the micro-cracks and the elastic strain energy released are increasingly large, resulting in micro-cracks along the surface of the glass and the thickness of the direction of value-added generation of the branch to form a new surface. The formation of ribbed median cracks inside the glass, also known as ‘Wallner ripples (WR)’, describes when the surface of the glass is roughened in the shape of a corrugated shell to absorb more elastic strain energy, forming lateral cracks. Crack extension is generally classified into three basic modes in the theory of fracture mechanics of brittle materials: type I cracks (open), type II cracks (slip-open), and type III cracks (tear-open), as shown in Figure 2. Crack extension when the scribing wheel cuts glass is generally a type I crack, found with Equation (2):

σ_{m} = σ [1 + 2 {(\frac{r}{ρ})}^{\frac{1}{2}}]

(2)

The stress at the crack tip is determined by a combination of the length of the crack (r) and the radius of curvature at the end (ρ); (σ_m) is the maximum stress at the tip of the crack, and (σ) is the externally applied stress.

The nature of the glass fracture process can be regarded as a process of internal stress change. Weller pointed out in the concept of glass stress distribution in the thickness [25] that glass after annealing of the original piece of internal pressure followed a ‘compressive stress—tensile stress—compressive stress (CTC)’ distribution, and that the essence of the glass fracture is the micro-cracks in the surface layer of compressive stress sprouting median cracks, used in the outside world to make the glass stress fracture uniform. Anton et al. [26] pointed out that the expansion of glass micro-cracks under a compressive stress layer is difficult and uncontrollable. Different shapes of micro-crack tips also affect the distribution of stress [27]. Accordingly, Schula et al. [28] assumed that the surface behind the micro-crack tip is always free of any external constraints, and deduced the stress component at any point in the micro-crack tip region to give the stress distribution equation of the glass crack in Equation (3), which is the ‘Irwin’s near-field solution’:

\{\begin{array}{l} σ_{xx} \\ σ_{yy} \\ σ_{xy} \end{array}\} = \frac{K_{1}}{\sqrt{2 π r}} \{\begin{array}{l} \cos (θ / 2) [1 - \sin (θ / 2) \sin (3 θ / 2)] \\ \cos (θ / 2) [1 + \sin (θ / 2) \sin (3 θ / 2)] \\ \cos (θ / 2) \sin (θ / 2) \cos (3 θ / 2) \end{array}\} + \frac{K_{1}}{\sqrt{2 π r}} \{\begin{array}{l} - (ρ / 2 r) \cos (3 θ / 2) \\ (ρ / 2 r) \cos (3 θ / 2) \\ - (ρ / 2 r) \sin (3 θ / 2) \end{array}\}

(3)

σ_xx, σ_yy, and σ_xy denote the lateral stress, longitudinal stress, and shear stress components at the crack tip in polar coordinates, respectively; K₁ is the equivalent stress intensity factor; r is the crack length, ρ is the radius of the curvature at the crack tip; and θ is the crack tip angle. The equations provide a method to calculate the stress field near the crack tip at different geometries for predicting the crack extension behavior.

3. Experiment Setup

3.1. Experiment Equipment

Figure 3 shows the test setup for glass scribing. The test was realized by modifying a surface grinder (M7130, Hangzhou Machine Tool Group Co., Ltd., Hangzhou, China) to cut the glass substrate. The cutting tool was obtained by assembling an aluminum alloy substrate with a notch of 10 mm width and threaded holes of 5 mm diameter, vertically distributed for the attachment of the scabbard. A fixed-size metal scribing wheel (HRC65 carbide scribing wheel, Zhejiang Sheng drill Hardware Co., Ltd., Taizhou, China) was mounted on the scabbard, the diameter of the scribing wheel was 2.5 mm, the hole diameter was 0.8 mm, and the thickness was 0.65 mm. Thin glass sheets (Float soda-lime glass, Luoyang Shangzhuo Technology Co., Ltd., Luoyang, China) with a thickness of 0.55 mm were selected and fixed to the mobile platform using a high-strength AB structural adhesive (6005 Epoxy Resin AB Glue, Elpida, Dongguan, China). The moving platform was driven by a synchronous belt slide module (LH4572, Beijing Time Mingwei Technology Co., Ltd., Beijing, China) and controlled by a controller to provide a horizontal travel speed of up to 400 mm/s. A force gauge (KISTLER 9272, KISTLER Instrumented AG, Winterthur, Switzerland) was fixed to the underside of the moving platform by an electromagnetic suction cup to ensure that the vertical pressure was constant during the cutting process.

The experiment cut-and-break process is shown in Figure 4a, where the scribing wheel scratches across the glass surface to create a median crack of a certain depth, and then using a splitting knife, pressure is applied perpendicularly from the back of the cut line of the glass to create a torque, which extends the median crack along the thickness of the glass to the bottom of the glass to separate the glass. Figure 4b shows a schematic diagram of the cutting crack morphology: the cutting produces a cutting micro-crack depth of d; the median crack is perpendicular to the glass surface and extends in the direction of the cutter wheel trajectory, with a length from the crack tip to the glass surface of l, and there is a deflection angle between the median crack and the cross-section. The lateral cracks are formed on the glass surface and are perpendicular to the cutting traces, and the spacing between the two tips of the lateral cracks is w. Figure 4c shows a schematic diagram of the morphology of the lobed cross-section, and the cut cross-section consists of three parts: the micro-cracks part, the median crack part, and the breaking part.

3.2. Experiment Parameters

Under the action of the cutting force F, the blade wheel cuts the glass substrate at speed v. In the contact area, the cutting force causes internal micro—cracks and chips in the glass, with the micro-crack depth d close to the cutting depth. Meanwhile, numerous transverse cracks, measured by width w, form around the scratch, affecting the strength of the post-cutting fracture surface and needing control during cutting. At the blade wheel’s top angle, two forces from its structure act on the bottom material, creating tensile stress that forms a median crack, which extends to a depth l along the glass thickness. In the subsequent scribing stage, the median crack further expands under the scribing torque until it penetrates the entire thickness. The median crack may deviate from the thickness direction during its formation and expansion, causing an offset angle α between the fracture surface and the thickness direction, thus affecting the grinding removal volume in later processes, and requiring control at this stage. The single-factor method was used in the experiments. The effects of the scribing wheel angle, scribing force, and scribing speed on the lateral crack w, median crack l, scribing depth d, and cross-section deflection angle α were investigated. Table 1 lists the detailed testing parameters. The range of the scribing wheel angle was 90–140°, with a step of 10°. The scribing force ranged from 10–30 N, with a step of 2 N. Under the interaction of wheel angle and scribing force, the scribing force was constant at 200 mm/s. Totally, 66 tests were conducted. After obtaining the optimal combination of scribing wheel angle and scribing force, the cutting test was carried out to investigate the effect of different scribing speeds on cutting quality.

At the end of the test, a laser confocal microscope (OLYMPUS LEXT OLS4100, Tokyo, Japan) was used to observe the crack morphology and measure the crack size. To ensure the accuracy and reliability of the experiments, each parameter was measured three times at different locations to calculate the average value.

4. Results and Discussions

4.1. Scribing Surface Topography

Figure 5 is an intensity image from a laser confocal microscope at 50× magnification. It gives the transverse crack patterns produced at 90°, 120°, and 140° cutter wheel angles at a cutting pressure of 20 N. These three angles were chosen because they are the most representative of the crack patterns produced by the cutter wheel. When the knife wheel angle is in the range of 90–110°, the cutting produces obvious lateral cracks, which are the most intensive around the scribing wheel traces to add value along the vertical cutting traces and produce branches, so that the traces are surrounded by a rough corrugated shell shape. This is because the small angle of the scribing wheel produces a larger CESF during cutting, which produces a scraping-cracking phenomenon (SCP), generating a large amount of glass debris, and the SCP increases with the increase in scribing force. Under the same scribing force, as the angle of the scribing wheel increases, it can be observed that the lateral cracks gradually extend from the surface to the subsurface, and the SCP produces less debris. This is because as the angle of the scribing wheel decreases, the direction of the CESF gradually converges to the inside of the glass, forming lateral cracks distributed on the subsurface of the glass. The cutting kerf for the scribing wheel angle in the range of 120–140° is narrow and straight, and no lateral cracks or chips are observed. This is because as the scribing wheel angle continues to increase, the CESF is not sufficient to produce lateral cracks or chips.

At a 90° scribing wheel angle and 20 N scribing force, the lateral cracks have a rough corrugated shell shape, accompanied by debris and chipping production, as shown in Figure 5a. Figure 5b shows that at the scribing wheel angle of 120° and scribing force of 20 N, the cutting micro-cracks are uniform and straight, and no lateral cracks or debris are produced. As shown in Figure 5c, it should be noted that at a 140° scribing wheel angle and 30 N scribing force, the glass was partially broken near the cutting mark due to excessive scribing force, resulting in edge bursting. The analysis shows that under the premise of ensuring that no lateral cracks are produced, the scribing wheel should be selected at as small an angle as possible and matched with as low a scribing force as possible to minimize damage to the glass surface.

4.2. Cross-Section Topography Analysis

After scribing the glass using a wheel, the glass was broken using by the bar, and the sectioned sample was rotated 90 degrees and fixed vertically under a confocal microscope to observe the cross-section. Figure 6 gives a cross-section of some of the cut and broken glass. The wheel angles and force values (e.g., (90°, 3 N); (90°, 10 N); (110°, 10 N); (120°, 20 N)) were selected for discussion because these parameter combinations showed significant and representative effects in the experiments, which can fully reflect the characteristics and trends of the research objects and at the same time avoid redundancy due to the selection of too many parameters in the analysis, ensuring the simplicity and efficiency of the study. The micro-cracks formed during scribing produce median cracks in the glass thickness direction due to the concentration of stress at the tip. A breaking part is formed below the median cracks after slicing. Figure 6a gives the cross-section at a scribing wheel angle of 90° and a scribing force of 3 N. Due to the insufficient scribing force, the median crack has not been formed yet, and the glass substrate cracks under the shear stress, resulting in edge cracks. As the scribing force increased, median cracks of constant depth accompanied by WR were observed below the plastic cut area, as shown in Figure 6b. The presence of glass debris at the edge of the cross-section when the scribing wheel was cut at an angle of 90° was due to the presence of a scraping-cracking phenomenon on the glass surface at this angle, which resulted in dense and large-sized lateral cracks. In Figure 6c, at a scribing wheel angle of 110° and a scribing force of 10 N, it is observed that there is a clear internal crack on the median crack and the brittle fracture region shows an uneven state. This is caused by the change in the direction of the cutting component force at the scribing wheel angle of 110° causing lateral cracks to form on the subsurface of the glass. At a scribing wheel angle of 120° and a scribing force of 20 N (Figure 6d), the median crack is flat and straight, with no obvious defects on the cross-cross-section, and the WR is clear, continuous, and evenly distributed. With the increase in scribing wheel angle and scribing force, as shown in Figure 6e,f, the chipping phenomenon occurs at the cutting edge, and the median crack exists with a messy and rough WR, intermingled with chipping and glass debris. The analysis shows that a smooth and neat WR can be obtained within the median crack, which is an important characterization for judging the quality of the glass cross-section.

4.3. Effect of Scribing Force and Scribing Wheel Angle on Crack Size

The results of the cutting surface/cross-section analyses show that the scribing wheel angle and scribing force are the key factors affecting the quality of glass cutting, and the selection of a suitable scribing force interval is particularly important for the quality control of glass substrate cutting. The effects of scribing force and scribing wheel angle on the size of the lateral cracks are given in Figure 7: the width of the lateral cracks shows an increasing trend with the increase in scribing wheel angle and scribing force, and is significantly divided into two regions as follows: in the interval of 90–110° of the scribing wheel angle, the lateral cracks are characteristically obvious, and the size is larger than 100 μm; in the interval of 120–140° of the scribing wheel angle, the lateral cracks are controlled in the range of 50 μm. Lateral cracks are controlled within 50 μm.

The slope was found for each point of the lateral crack, and the effect of the scribing force on the rate of expansion of the lateral crack is given in Figure 8. With the increase in scribing force, the lateral crack extension rate shows a tendency of increasing and then decreasing, and this process is divided into three stages [29,30]. In the micro-crack nucleation stage, the glass surface of the micro-cracks begins to sprout and propagate due to stress concentration; because the initial micro-crack size is small, the crack expansion of the overall damage to the material has a smaller impact, so the crack width increases more slowly. In the micro-crack propagation stage, sprouted micro-cracks begin to expand along the direction of the stress concentration, and the rate increases. In the crack connection stage, with the further increase in scribing force, the expansion of the micro-cracks leads to cracks between the interconnection and the formation of a larger crack network, resulting in local material damage due to the formation of chipping. The analyses show that the cutting micro-cracks were in the micro-crack nucleation stage when the scribing force was within 22 N and the scribing wheel angles were 120° and 130°, which caused little damage to the glass surface and did not produce lateral cracks and chipping.

Figure 9 demonstrates the effects of the scribing force and scribing wheel angle on the cutting micro-crack depth and median crack length. The results show that under the same scribing force, the depth of cuts shows an increasing trend with an increase in the scribing wheel angle, and the median crack shows a decreasing trend, which indicates that adjustment of the scribing wheel angle adjusts the relative action of plastic deformation and elastic deformation on the glass [31]. The 90–110° scribing wheel angle is more likely to produce longitudinal cracks, but the actual depth of cut is shallower due to the scraping-cracking phenomenon that occurs on the surface of the glass or due to cracking on the subsurface, which produces lateral cracks and strips of broken glass at the edges of the cutting knife marks. As the scribing wheel angle increased to 120–140°, the micro-cracks depth increased. At the same scribing wheel angle, the median crack and micro-crack depth increased with increasing scribing force. It should be noted that compared to the 130° and 140° scribing wheels, at a scribing wheel angle of 120°, the micro-crack depth produced by budding the same depth of median cracks was lower, and less scribing force was required, which is more in line with the principles of selecting the scribing wheels and scribing forces in the above results.

4.4. Effect of Cutting Parameters on the Stress Field at the Crack Tip

The analysis of the cutting surface, cross-section morphology, and crack size shows that the fracture process of the crack is essentially the process of internal stress change. Figure 10, based on the glass fracture crack stress distribution equation, shows the effect of different cutting parameters on the crack tip stress distribution, taking the glass substrate three-dimensional type I crack tip stress intensity factor size as K_I = 78.9 MPa·m^1/2 [32]. If the effect of crack tip curvature is not considered, Figure 10a gives the effect of scribing wheel angle on crack tip stress at a 16 N scribing force. As the scribing wheel angle increases, the lateral stress, longitudinal stress, and shear stress around the crack tip show a decreasing trend, implying a corresponding decrease in lateral and longitudinal cracks. Figure 10b gives the effect of change in scribing force on crack tip stress at a 120° scribing wheel angle. As the scribing force increases, the stress around the crack subsequently increases, leading to an increase in crack size. If chipping occurs during cutting, the crack tip changes from sharp to having a certain curvature α. Compared to Figure 11a, at 140°, the longitudinal and tangential stresses further decrease while the transverse stresses increase. The analyses show that the fracture process of the crack is closely related to the scribing force and internal stress changes, and that chipping should be avoided during cutting.

4.5. Effect of Cutting Parameters on Cross-Section Deflection Angle

Because the internal stress of glass follows a CTC distribution, the essence of glass fracture is that micro-cracks on the surface layer of compressive stress form median cracks and make the glass fracture uniformly through the external torque. If the length of the median crack generated during cutting is not long enough to exceed the compressive stress layer, it will result in the cross-section deflection angle, as the cut cross-section is not perpendicular to the median crack. As shown in Figure 11a, the flatness of the cut cross-section is assessed by measuring the angular deviation between the median crack portion of the cross-section and the fractured portion of the lobe [33]. As shown in Figure 11b, when the cross-cross-section deflection angle is 1.2°, the cross-section heights are uniformly distributed and the cross-section articulations are smooth. On the contrary, at a cross-section deflection angle of 5.6°, the height maps show stratification (Figure 11c) and angular deflection at the cross-section articulation.

The effect of different cutting parameters on the angular deviation is given in Figure 12. As shown in Figure 12a, the angular deviation is in the range of 3–6° for scribing wheel angles of 90–110°, with no obvious distribution trend. This is caused by the obvious scraping-cracking phenomenon and the high stress on the horizontal sides (Figure 12a). The effect of the scribing force on the median crack and cross-section deflection angle at 120–140° is given in Figure 12b–d, where the cross-section deflection angle shows a tendency of decreasing and then increasing with increasing scribing force. This is because when the scribing force is small, the length of the median crack sprouting is lower than the thickness of the compressive stress layer of the glass itself, the expansion of the crack compressive stress layer shows irregularity, and the cross-section deflection angle is in the range of 3–6°. With the median crack sprouting to about 20% of the glass thickness, the crack sprouting is more concentrated and controllable, and the cross-section deflection angle decreases to within 2°. However, as the scribing force increases further, the resulting chipping phenomenon leads to instability in the cracking process, and the cross-section deflection angle rises again to over 3°, with an increasing trend. This is in line with existing studies that show that the distribution of soda–lime glass at different formulation ratios roughly follows the 0.2t–0.6t–0.2t law [34]. Therefore, it should be cut so that the median crack exceeds 20% of the glass thickness.

4.6. Influence of Scribing Speed

The effect of the scribing speed on lateral crack, longitudinal crack, depth of cut, and deflection angle is given in Figure 13. Experiments were carried out at a scribing wheel angle of 120°, a scribing force of 16 N, and a scribing speed of 100 mm/s–400 mm/s. The lateral crack length varied within 2 μm, the cross-section deflection angles were all below 2°, the median crack length varied within 10 μm, and the micro-crack depth varied within 5 μm. The analyses show that within the range of test speeds, the variation in the scribing speeds does not negatively affect the quality of the cut, providing flexibility for parameter optimization of the cutting process.

5. Conclusions

In this study, micro-cracks in the glass scribing process were systematically assessed through discussing the effect of the scribing wheel angle, scribing force, and scribing speed. The conclusions are as follows:

(1): Lateral cracks increase as the scribing wheel angle and the scribing force increase. A scribing wheel angle in the range of 90–110° produces a planning and scraping effect, so that the scribing marks are rough, corrugated, and shell-like. When the scribing wheel angle is in the range of 120–140°, the glass surface damage is small and the scribing wheel mark is narrow and straight, without lateral cracks and debris generation.
(2): The median crack and scribing depth showed an increasing trend with increasing scribing force. However, as the scribing wheel angle increased, the scribing depth showed an increasing trend and the median crack showed a decreasing trend.
(3): With the increase in the scribing wheel angle, the stress around the crack tip shows a decreasing trend. With the increase in the scribing force, the stress around the crack increases. The chipping phenomenon occurs when the scribing wheel angle or scribing force is too large. The micro-crack tips change from sharp to having a certain curvature, so that the lateral cracks increase in stress and the surface breaks.
(4): When the scribing wheel angle is in the range of 90–110°, the cross-section deflection angle is in the range of 3–6°. When the scribing wheel angle is in the range of 120–140°, the cross-section deflection angle increases and then decreases with the increase in scribing force. The cross-section deflection angle is the smallest when the median crack sprouts to about 20% of the glass thickness.
(5): The optimized parameters based on the scribing quality are a 120° scribing wheel angle and 20 N scribing force. The scribing speed in the range of 100–400 mm/s does not affect the scribing quality.

Author Contributions

Conceptualization, and methodology, H.K.; validation, resources, and data curation, L.H.; investigation and writing—original draft preparation, D.L.; software, writing—review and editing, J.L.; visualization, supervision and funding acquisitionn, Y.L.; project administratio, J.G. All authors have read and agreed to the published version of the manuscript.

Funding

This research was supported by the Shanxi Province Science and Technology Major Project (Granted No. 202301150401007) and Taiyuan City Science and Technology Major Project (Granted No. 2024TYJB0128).

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

The original contributions presented in this study are included in the article. Further inquiries can be directed to the corresponding author.

Conflicts of Interest

Author Dawei Li, Jinzhu Guo, Liyong Huang and Yao Liu were employed by the company CETC Fenghua Information Equipment Co., Ltd. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

References

Cao, L.N.; Cai, K.X. Study on Cutting Technology of Large Size Liquid Crystal Glass. Electron. Process Technol. 2018, 39, 231–234. [Google Scholar]
Li, Y.L.; Zhang, F.H. Discussion on the technological parameters of TFT-LCD chip cutting. Chin. J. Liq. Cryst. Disp. 2006, 21, 43–47. [Google Scholar]
Lee, C.H.; Wu, K.C.; Chiang, K.N. A Novel Acceleration-Factor Equation for Packaging-Solder Joint Reliability Assessment at Different Thermal Cyclic Loading Rates. J. Mech. 2016, 33, 35–40. [Google Scholar] [CrossRef]
Allaire, R.A. Fracture Analysis, a Basic Tool to Solve Breakage Issues; Shizuoka Technical Center: Shizuoka, Japan, 2004. [Google Scholar]
An, Q.L.; Ming, W.W.; Chen, M. Experimental Investigation on Cutting Characteristics in Nanometric Plunge-Cutting of BK7 and Fused Silica Glasses. Materials 2015, 8, 1428–1441. [Google Scholar] [CrossRef] [PubMed]
Cheng, H.C.; Li, K.H.; Shih, C.Y.; Chen, W.H. Characterization of the Flexural Strength and Fatigue Life of Ultrathin Glass After Dicing. IEEE Trans. Compon. Packag. Manuf. Technol. 2018, 8, 2213–2221. [Google Scholar] [CrossRef]
Murata, T.; Miwa, S.; Yamazaki, H.; Yamamoto, S. Suitable Scribing Conditions for AMLCD Glass Substrates. SID Symp. Dig. Tech. Pap. 2003, 34, 374–377. [Google Scholar] [CrossRef]
Li, H.Q.; Wang, T.; Li, D.W. Research on cutting and breaking technology of TP cover glass. Mech. Manag. Dev. 2019, 34, 59–61. [Google Scholar]
Luo, S.Y.; Tsai, Y.Y.; Chen, C.H. Studies on cut-off grinding of BK₇ optical glass using thin diamond wheels. J. Mater. Process. Technol. 2006, 173, 321–329. [Google Scholar] [CrossRef]
Li, J.S.; Guo, B.F.; Zhang, B.D. Problems and Their Solutions in Cutting Process for Liquid Crystal Glass Substrate. Glass Enamel Ophthalmic Opt. 2020, 48, 19–21+24. [Google Scholar]
Zhou, A.F.; Mu, M.Q. Improving the Precision Control of Glass Substrate Cutting. Glass 2018, 45, 36–39. [Google Scholar]
Pei, X.D.; Su, B.H.; Xue, J.W.; Wang, L.L. Study of the Influence of Cutting Parameters on LCD Machining Quality. Infrared 2018, 39, 30–35. [Google Scholar]
Zhou, B. Analysis and Optimization Measures of LCD Substrate Glass Cutting Process. Value Eng. 2015, 34, 91–92. [Google Scholar]
Ono, T.; Tanaka, K. Effect of scribe-wheel dimensions on the cutting of AMLCD glass substrate. J. Soc. Inf. Disp. 2001, 9, 87–94. [Google Scholar] [CrossRef]
Li, X.; Wang, X.; Jiang, Y.; Wan, Q.; Chen, D. Analysis of Cutting Process for 0.20~0.33 mm Ultra-thin Float Glass. Glass Enamel 2019, 47, 1–6. [Google Scholar]
Chen, R.; Zhou, C.; Xie, J.; Chen, Z.J. Experimental study on rolling and brittle fracture to single crystal silicon and sapphire by diamond cutter wheel. Opt. Precis. Eng. 2022, 30, 702–709. [Google Scholar] [CrossRef]
Wu, F.; Liu, Q.; Wang, H.; Liu, F.Q.; Luo, C.J.; Li, M.H. Cutting Wave Defect Improvement with the Scribing Procedure. Optoelectron. Technol. 2022, 42, 58–63. [Google Scholar]
Swain, M.V.; Metras, J.C.; Guillemet, C.G. A deformation and fracture mechanics approach to the scoring and breaking of glass. J. Non-Cryst. Solids 1980, 38–39, 445–450. [Google Scholar] [CrossRef]
Pan, C.T.; Hsieh, C.C.; Su, C.Y.; Liu, Z.S. Study of cutting quality for TFT-LCD glass substrate. Int. J. Adv. Manuf. Technol. 2008, 39, 1071–1079. [Google Scholar] [CrossRef]
Zhang, X.H.; Cai, K.X.; Zheng, Y. LCD glass marking machine cutting process improvement. Shanxi Electron. Technol. 2015, 3, 25–26. [Google Scholar]
Li, Q.; Wang, B.Z.; Zhang, Z.Y.; Cao, H.Y.; Wu, D.; Zheng, Q. Causes and Control of Glass Adhesion Defects in TFT-LCD Substrate Glass. Glass 2020, 47, 27–31. [Google Scholar]
Wu, F.G.; Li, R.; Lv, F.Z.; Liu, F.Q. TFT-LCD industry liquid crystal glass cutting Dummy control and improvement. Plant Maint. Eng. 2021, 10, 68–72. [Google Scholar]
Liu, Q.; Zhang, F.H.; Li, Y.L. Effect of Optimized Scribing and Breaking Conditions on Cleland Module Process for TFT-LCD Glass Substrates. Adv. Disp. 2006, 09, 43–46. [Google Scholar]
Song, X.H.; Fan, Z.X. The Defects of LCD Scribing and Breaking, and it’s Solution. Adv. Disp. 2010, 11, 18–22. [Google Scholar]
Weller, R. Three-Dimensional Photoelasticity Using Scattered Light. J. Appl. Phys. 1941, 12, 610–616. [Google Scholar] [CrossRef]
Hödemann, S.; Möls, P.; Kikas, J.; Anton, J. Scattered laser light fringe patterns for stress profile measurement in tempered glass plates. Glass Technol. Eur. J. Glass Sci. Technol. Part A 2014, 55, 90–95. [Google Scholar]
James, M.N. Residual stress influences on structural reliability. Eng. Fail. Anal. 2011, 18, 1909–1920. [Google Scholar] [CrossRef]
Chula, S. Charakterisierung der Kratzanfälligkeit von Gläsern im Bauwesen; Springer Vieweg: Berlin/Heidelberg, Germany, 2015. [Google Scholar]
Li, Y.T.; Yan, J.H.; Gao, J.J.; Wang, J. Quality Control of Cutting and Edge Snapping for Ultra Thick and Ultra Long Float Glass. World Build. Mater. 2024, 45, 65–68. [Google Scholar]
Wang, J.; Zhang, Y.F.; Ma, J.Y.; Li, J.X.; Zhang, Z. Microcrack Nucleation and Propagation Investigation of Inconel 740HAlloy Under in Situ High Temperature Tensile Test. Acta Metall. Sin. 2017, 53, 1627–1635. [Google Scholar]
Zhao, A.H.; Yu, J.L. Microscopic damage evolution model of quasi-brittle materials. J. Tsinghua Univ. (Sci. Technol.) 2000, 5, 88–91. [Google Scholar]
Wu, X.; Lei, C.J.; Lv, Y.C.; Li, J.G.; Li, Y.; Zhou, H.K. Scoring Angle, Pressure, Speed Effect of Glass Median Crack and Breakout Quality and Improved Scoring Method Based on Median Crack Analysis. Glass 2022, 49, 1–10+19. [Google Scholar]
Liu, J.S.; Wang, P.Y. Numerical calculation of stress intensity factor of 3D type I crack tip based on Abaqus. In Proceedings of the Beijing: The 25th Annual Conference of Beijing Force Society, Beijing, China, 6 January 2019. [Google Scholar]
Hundhammer, I.; Lenhart, A.; Pontasch, D.; Weissmann, R. Stress measurement in transparent materials using scattered laser light. Glass Sci. Technol. 2002, 75, 236–242. [Google Scholar]

Figure 1. Schematic of scribing force.

Figure 2. Fracture modes.

Figure 3. Experimental setup.

Figure 4. Schematic diagram of scribing and breaking process. (a) Glass scribing and breaking process, (b) scribing line, and (c) cross-section view.

Figure 5. Effect of different scribing wheel angles and scribing forces on the surface topography of glass substrate. (a) θ = 90° and F = 20 N, (b) θ = 120° and F = 20 N, (c) θ = 140° and F = 30 N.

Figure 6. Cutting cross-section morphology of glass by using (a) θ = 90° and F = 3 N, (b) θ = 90° and F = 10 N, (c) θ = 110° and F = 10 N, (d) θ = 120° and F = 20 N, (e) θ = 140° and F = 20 N, (f) θ = 140° and F = 30 N.

Figure 7. Effect of scribing force and scribing wheel angle on lateral crack size.

Figure 8. Effect of scribing force on the propagation rate of lateral cracks.

Figure 9. Effect of scribing force and scribing wheel angle on cutting micro-crack depth and median crack length (line graph—median crack; bar graph—micro-crack depth).

Figure 10. Stress variation at the crack tip with different (a) scribing wheel angle and (b) scrining force.

Figure 11. Height cloud map and 3D views of glass cross-sections at different cutting parameters. (a) surface profile for deflection angle measurement, (b) deflection angle at θ = 130° and F = 16 N, and (c) deflection angle at θ = 90° and F = 10 N.

Figure 12. (a) Effect of scribing force and scribing wheel angle on cross-section deflection angle at 90–110°; (b) 120°, (c) 130°, (d) 140°; effect of scribing force on median crack and cross-section deflection angle.

Figure 13. Effect of scribing speed on lateral cracks, median cracks, micro-crack depths, and cross-section deflection angle. (a) lateral crack and deflection angle, (b) microcrack depth and median crack.

Table 1. Experiment parameters.

Parameters	Value
Scribing wheel angle (°)	90, 100, 110, 120, 130, and 140
Scribing force (N)	10, 12, 14, 16, 18, 20, 22, 24, 26, 28, and 30
Scribing speed (mm/s)	100, 150, 200, 250, 300, 350, and 400

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

© 2025 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).

Share and Cite

MDPI and ACS Style

Li, D.; Li, J.; Kong, H.; Guo, J.; Huang, L.; Liu, Y. Investigation of Ultra-Thin Glass Scribing Mechanism. Coatings 2025, 15, 275. https://doi.org/10.3390/coatings15030275

AMA Style

Li D, Li J, Kong H, Guo J, Huang L, Liu Y. Investigation of Ultra-Thin Glass Scribing Mechanism. Coatings. 2025; 15(3):275. https://doi.org/10.3390/coatings15030275

Chicago/Turabian Style

Li, Dawei, Jiahao Li, Huaye Kong, Jinzhu Guo, Liyong Huang, and Yao Liu. 2025. "Investigation of Ultra-Thin Glass Scribing Mechanism" Coatings 15, no. 3: 275. https://doi.org/10.3390/coatings15030275

APA Style

Li, D., Li, J., Kong, H., Guo, J., Huang, L., & Liu, Y. (2025). Investigation of Ultra-Thin Glass Scribing Mechanism. Coatings, 15(3), 275. https://doi.org/10.3390/coatings15030275

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Menu