Abstract
This study systematically examines the wear performance and machining efficiency of vacuum-brazed diamond-coated milling tools under optimized processing conditions. Two widely used metamorphic-origin marbles, Muğla White (MW) and Uşak White (UW), were selected as workpieces. The experimental methodology involved milling operations on a CNC machining center, where spindle speed (6000–12000 rpm), feed rate (1500–3000 mm/min), and a constant depth of cut (1 mm) were selected as the key input process parameters. Machining performance was evaluated through surface roughness (Ra), cutting force (Fr), specific energy consumption (SE), and material removal rate (MRR), while wear mechanisms were characterized using Scanning Electron Microscopy (SEM) and digital microscopy on machined areas of 1 to 7 m2. Optimization analysis determined the ideal processing conditions to be 6000 rpm and 2281 mm/min for MW, and 12000 rpm and 3000 mm/min for UW, which were subsequently confirmed by validation experiments. Comparative findings indicated that the lower mechanical strength of MW marble resulted in greater material removal but increased tool wear (54%) compared to UW (47%). These results highlight the decisive influence of microstructural characteristics and operating parameters on tool wear, thereby contributing to the industry by offering guidelines for extending tool lifecycle and enhancing energy efficiency.
Keywords
Introduction
Globally renowned for its abundant natural stone reserves, Turkey holds a significant position in marble production, particularly for metamorphic-origin varieties extracted from various regions. These natural stones have traditionally been recognized for their exceptional strength, workability, and aesthetic appeal, offering a diverse range of applications in both architectural and artistic domains. Marbles, such as MW and UW, have historically been used to manufacture decorative products, flooring, wall coverings, sculptures, and interior designs owing to their unique properties. In this research, the physical, chemical, and mechanical characteristics of marble samples obtained from significant deposits in Turkey were comprehensively analyzed. Results shed light on the scientific and industrial application potential of these stones, presenting opportunities to drive advancements in related fields.1–3
The application of diamond coatings on milling tools using vacuum brazing method represents an innovative technique that significantly improves tool longevity and performance, particularly in the marble processing industry. Despite these advantages, vacuum-brazed diamond tools face several challenges, primarily associated with wear mechanisms and chip formation during machining. These limitations influence the tool lifespan, product quality, and overall cost, making them critical factors in determining production efficiency. Considering the high hardness and abrasive nature of natural stones, examining the impact of diamond coating methods on tool wear is crucial for ensuring efficient and high-quality stone machining. The milling process is integral to crafting unique natural stone designs, where boosting surface quality, minimizing tool wear, and enhancing tool performance are key to achieving cost-effective and successful manufacturing outcomes. Compared to carbide and electroplated tools, vacuum-brazed diamond tools stand out by offering reduced cutting forces, superior surface finishes, and prolonged tool life. Additionally, the properties of the stone and the size of the diamond grains significantly impact the tool performance. By optimizing machining parameters and selecting appropriate tools, improved outcomes can be attained.4–7
Critical parameters such as cutting force (Fr), average surface roughness (Ra), specific energy (SE), and material removal rate (MRR) play a pivotal role in optimizing milling performance, increasing efficiency, and ensuring cost-effectiveness in the processing of natural stone.8–12 The interplay between these factors heavily influences machining optimization strategies, as higher cutting forces and specific energy tend to accelerate the wear. At the same time, carefully optimized parameters can significantly mitigate wear rates and enhance tool service life.13,14
During marble milling, various wear types are commonly observed in electroplated diamond tools, predominantly in the edge and corner regions. These wear patterns stem from excessive stress, repetitive impacts, and vibrations, ultimately leading to diamond grain fracture and fragmentation. Hard rock particles introduce macro- and micro-fractures in diamond grains, accelerating material loss. Consequently, this damage causes the tool to become dull over time, thereby decreasing its cutting performance.15–19
The processing parameters and mineralogical characteristics of the machined stone are critical factors that influence wear mechanisms and tool longevity. The key parameters in granite milling include cutting speed, feed rate, and diamond grain size. Increasing the cutting speed generally helps reduce edge chipping, although this improvement has a threshold beyond which the effects become unpredictable. Moreover, lower feed rates and finer diamond grain sizes help mitigate edge chipping and achieve higher surface quality.20–23 The mineralogical composition and grain size of granite significantly impact average surface roughness and chipping tendencies, with individual minerals contributing distinctively to the resulting surface finiş.20,24,25
The metal matrix supporting the diamonds undergoes wear, most notably in the front segment of the tool, where dynamic loads and impacts are most intense, leading to the detachment of diamond grains over time.26,27 At high material removal rate (MRR) levels, increasing tool wear or unstable process conditions may limit or reverse these effects.28,29
While recent studies have emphasized advanced fabrication techniques like CVD coatings and uncertainty analysis to enhance data reliability,30,31 the application of these methodologies to heterogeneous natural stones remains complex. In the context of stone machining, milling cutters are predominantly categorized as carbide-coated, electroplated, or vacuum-brazed. Previous research has extensively documented the limitations of carbide tools, noting their rapid wear and inability to maintain surface quality in abrasive conditions. Similarly, studies on electroplated diamond tools15–19 have characterized wear patterns, specifically highlighting premature diamond grain detachment due to plating defects.
Despite the extensive literature on granite milling and electroplated tools,20–25 there is a significant research gap regarding the performance of vacuum-brazed (VB) diamond tools specifically for metamorphic-origin marbles. Most existing research focuses on either tool life or surface quality independently, often neglecting the holistic correlation between specific energy consumption (SE), cutting forces, and the progressive wear mechanisms of VB tools. Although VB technology offers superior grain retention and chip clearance compared to traditional methods,7,32,33 the optimization of milling parameters for soft-to-medium hard marbles like MW and UW has not been sufficiently explored. Therefore, this study aims to bridge this gap by systematically analyzing the wear behavior and energy efficiency of VB tools under optimized conditions.
The objective of this research is to identify optimal milling parameters and analyze tool wear behavior under selected conditions. Since tool wear significantly affects tool life, product quality, and production cost, the specific aims of this study are presented as follows:
To identify optimal milling parameters and analyze tool wear behavior under selected conditions. To optimize machining parameters specifically for the structural characteristics of marble. To improve machining efficiency, reduce costs, and extend tool life by mitigating the adverse effects of tool wear.
Materials and methods
Experimental process
In this study, a four-axis CNC vertical machining center (Megatron–Megastone 2040, Türkiye), widely utilized in the natural stone processing industry, was selected for the experiments (Figure 1). The machine is equipped with a 9 kW (12 HP) spindle motor capable of a maximum rotational speed of 24,000 rpm and a processing speed of up to 24,000 mm/min. Additionally, the axis feed rate can reach a maximum of 80,000 mm/min, ensuring high-speed operational capability. Water was used as the coolant during the milling operations, with a flow rate of 3 L/min. When cooling is not applied during marble machining, excessive heating occurs on the tool's working surface, leading to the burning of the diamond coating and tool material, rapid wear, and fracturing. Therefore, water cooling was used in all experiments.
Experimental process flow chart.
The experiments employed newly developed diamond-coated milling tools produced using vacuum-brazing technology. The total length of the tool was 60 mm, with a cutting part measuring 20 mm and a width of 6 mm. The diamond coating size falls within the range of 25/35 mesh. The composition of the coating material is kept as a trade secret by the manufacturer.
The experimental measurement chain and data acquisition process are schematically illustrated in Figure 1. A custom-designed dynamometer system, integrated with eight ESIT BS model load cells mounted on the test bench, was employed for force measurement. Four of these sensors were configured to monitor vertical forces (Fz), while the remaining four were arranged to detect horizontal forces (Fx, Fy). The tangential cutting force (Ft) was calculated using the equation The experimental measurement chain and data acquisition process are schematically illustrated in Figure 1. A custom-designed dynamometer system, integrated with eight ESIT BS model load cells mounted on the test bench, was employed for force measurement. Four of these sensors were configured to monitor vertical forces (Fz), while the remaining four were arranged to detect horizontal forces (Fx, Fy). The tangential cutting force (Ft) was calculated using the equation
6
The mineralogical structure, along with the crystal sizes and morphologies of the marble samples, was assessed via thin-section analyses conducted using a Nikon Eclipse 2V100POL electron microscope. The mineralogical and petrographic properties of the marbles are detailed in Table 1, while microstructural images of the thin sections are illustrated in Figure 2. The mineralogical compositions and microstructural characteristics of the Muğla White (MW) and Uşak White (UW) marble samples were examined using polarized light microscopy, as shown in Fig. 2. The MW sample (Left) exhibits a heteroblastic texture characterized by coarse, irregular calcite grains with distinct cleavage planes and relatively loose intergranular boundaries. In contrast, the UW sample (Right) displays a homeoblastic texture with a finer, more uniform grain size distribution. The grains in the UW sample demonstrate a compact mosaic structure with tighter interlocking boundaries compared to the MW sample. These microstructural variations are critical indicators of the physical and mechanical heterogeneity between the two metamorphic origins.
Mineralogical microscope images (left: mw, right: uw)
Mineralogical and petrographic properties of marbles.
The chemical composition of the marble samples was determined using X-ray Fluorescence (XRF) spectrometry. XRF is a non-destructive analytical technique widely used to quantify the elemental composition of geological materials by measuring the secondary X-rays emitted from a sample when excited by a primary X-ray source. The major oxide components obtained from this analysis are presented in Table 2, providing insight into the mineralogical characteristics of the studied marbles.
XRF analysis results.
Finally, to evaluate the structural and strength attributes of the marble samples, a series of tests was administered. Physical characteristics–including specific gravity, water absorption, and porosity—were measured, whereas mechanical tests assessed compressive strength, flexural strength, and abrasion resistance. The aggregated results of the physical and mechanical evaluations are provided in Table 3.
Physical and mechanical analysis results.
A full factorial experimental design with two factors and four levels was employed to systematically evaluate the milling performance and the interactions between processing parameters, resulting in a total of 16 distinct experimental conditions (Table 4). The selection of fixed processing parameters—specifically cutting depth, cutting width, and diamond grit size (50/60 mesh)—was grounded in preliminary experimental trials and findings from our previous studies.6,7 Since these studies demonstrated that the selected conditions yield optimal performance regarding specific energy consumption, average surface roughness, and cutting forces, these values were maintained as constants in the current investigation to ensure experimental consistency and to isolate the effects of the variable parameters.
Processing parameters.
Processing parameters.
All measurement and calculation results obtained from the experimental studies are summarized in Table 5. An examination of the data presented in the table reveals that variations in machining parameters induce significant differences in cutting forces, surface quality, and energy consumption for both marble types (MW and UW). These experimental findings were utilized to analyze the effects of cutting parameters on machinability and to conduct comparative evaluations. Furthermore, the obtained results constitute the fundamental basis for characterizing the machining performance of the marbles and validating the accuracy of the developed regression models.
Machining measurement results of mw and uw marbles.
Machining measurement results of mw and uw marbles.
To model and forecast the process variables—including spindle speed and feed rate—alongside determining favorable machining conditions, Response Surface Methodology (RSM) was adopted. The constraints for the process parameters are listed in Table 6.
Parameter constraints.
Using the gathered experimental data, predictive functions were formulated to correlate the dependent and independent variables, facilitating the estimation of the system responses (Table 7).
Prediction functions for marbles.
Prediction functions for marbles.
A comprehensive analysis was performed by comparing the predicted results with the experimental data. Separate graphs were generated for the three key parameters (Fz, Ra, and SE) to display the relationship between the measured and predicted values. The data points in these graphs are distinguished by various colors, corresponding to different parameter values. Moreover, each graph includes the R2 coefficient, highlighting the statistical accuracy and reliability of the prediction models.
The estimated values derived from the prediction functions were compared with the actual measured values, and the correlation between them (R2) was examined. The results of this comparison are illustrated in Figure 3.
Validation of empirical models: analysis of actual and predicted distribution and coefficient of determination (R2) for mw and uw marbles.
The calculated R2 values for the marble samples revealed that the prediction models efficiently estimated the actual parameter values. Strong correlations (R2 > 0.8) were noted for parameters such as Fr, Ra, and SE. These results demonstrate the reliability and effectiveness of the developed prediction models in forecasting the process parameters during the machining of MW and UW marbles.
The optimal machining conditions obtained through optimization studies for MW and UW, along with the corresponding performance parameters under the ideal conditions, are summarized in Table 8.
Optimization results.
As a result of the optimization efforts, the most applicable process parameters for the AB and AG marbles were successfully determined. Table 8 outlines the forecasted performance indicators for spindle speed, feed rate, Fr, MRR, Ra, SE and desirability parameters. The high desirability values for both marble types imply that the chosen process conditions effectively met the various performance objectives.
The characteristics of tool wear during the machining of marbles using vacuum-brazed diamond milling tools were meticulously investigated under optimal milling conditions. The primary goal of this study was to assess the interaction between the tool and workpiece, documenting the wear mechanisms experienced in marbles with varying grain sizes, specifically Marble White (MW) and Uşak White (UW) marbles. Figure 4 illustrates a schematic depiction of the tool tip region alongside a microscopic image of the vacuum-brazed diamond milling tool. This visual also highlights the cutting and feed directions employed during processing, effectively explaining the chip removal mechanism for the MW and UW marbles. Furthermore, it provides a detailed overview of the contact mechanisms and chip formation patterns between diamond grains and the material.
Graphical representation of processing data obtained during the wear of mw and uw marbles.
A thorough evaluation of the wear condition of the diamond milling tool was carried out. Wear progression tests were conducted incrementally on areas of 1, 3, 5, and 7 m2 and analyzed through microscopic observation. t each stage of the experimental process, high-resolution images of the tool surfaces were acquired to analyze wear mechanisms and surface morphology. The microstructural analysis was conducted using a Scanning Electron Microscope (SEM) for detailed topographical examination. Additionally, macro-scale surface characterization was performed using a Dino-Lite digital microscope equipped with a 1.3-megapixel CMOS sensor, capable of providing magnification in the range of 20x to 90x. Separate milling tools were used to test the MW and UW marble specimens, and the data are summarized in Table 9.
Wear analyses by stages.
The state and surface characteristics of the diamond particles before wear were examined for both the MW and UW marbles (Tables 10 and 11). For the MW marble, the initial setup of the tool featured 32 active diamonds, providing a cutting surface area ratio of 24%. Post-wear analysis revealed that 26 diamonds showed evidence of wear, with the cutting surface area reduced to 11%, representing a total reduction of 54%. Observations highlighted degradation phenomena such as edge and corner wear, as well as fracturing and fragmentation, starting from the machining of 1 m2.
In comparison, the tool for UW marble incorporated 28 active diamonds, with an initial cutting surface area ratio of 17%. After the wear tests, it was observed that 23 diamonds experienced wear, leading to a decrease in the cutting surface area of 9%—a reduction of 47%. The wear mechanisms identified included edge and corner wear, as well as fracturing at 1 and 3 m2, with fragmentation becoming pronounced after the 3 m2 surface was machined.
During the milling process of MW marble, notable variations in the machining parameters were observed after processing 1 m2, primarily due to the effects of tool wear and diamond grain fragmentation. The Ra and SE values exhibited a continuous upward trend that was directly linked to the progressive fragmentation of the diamond grains. A substantial increase in SE was recorded between 3 and 5 m2, which correlated with alterations in the diamond cutting surface attributes.
The stability of Fr throughout the process demonstrates the superior diamond retention capability of the VBD technology. The sudden force drops typically observed in electroplated tools due to diamond pull-out were not present in this study. This is a testament to the success of VB technology. However, the increase in SE indicates that while the diamonds remained intact, they underwent dulling, causing a transition from the ‘cutting’ mode to the ‘ploughing’ mode. Consequently, the tool consumes more energy to perform the cutting operation.
In the case of UW marble, the Ra value exhibited a pronounced rise beyond 3 m2 in relation to cutting forces, accompanied by a significant increase in Fr after machining 5 m2. The Ra value increased consistently during the process. While SE value generally showed stability, an abrupt drop was detected at 5 m2 due to diamond fragmentation, followed by a sharp escalation thereafter.
To evaluate the particle size distribution of marble chips, analyses were performed using a particle size analyzer (Mastersizer 2000), which is capable of measuring particles within the range of 0.02 μm to 2000 μm. Commonly used metrics, such as D10, D50, and D90, were applied to analyze the data. These parameters are integral for characterizing materials, assessing the efficiency of milling and crushing processes, and evaluating the influence of various process parameters on the particle size distribution. D10 signifies that 10% of the sample's particles are smaller than this size, D50 represents the median size where 50% of the particles fall below it, and D90 indicates the size below which 90% of the particles are located. The respective results are detailed in Table 12. Additionally, the interactions between the tool and workpiece, as well as chip formation during marble machining with diamond-coated tools, are illustrated in Figure 5.
Chip mechanics and tool interaction in the machining of marbles with diamond milling tools.
SEM images for MW.
SEM images for UB.
Results of chip particle size analysis of marbles.
The particle size analysis graphs (Figure 6) demonstrate a multimodal distribution with distinct peaks detected at various size ranges, indicating non-uniform chip formation. These results suggest the presence of multiple mechanisms of mechanical fracture and detachment. Predominantly, large and medium-sized chips were observed during marble milling, as highlighted in Figure 7. Medium-sized chips were formed as the cutting tool progressively removed small fragments from the surface, whereas larger chips were generated during deeper cuts or abrupt loading, leading to substantial breakage. Medium- and large-sized chip particles are regarded as significant indicators of both the cutting tool's efficiency and degradation.
Analysis of particle size distributions of chips formed after machining (left: mw, right: uw).
Optical microscope images of surface morphology of marble chips (left: mw, right: uw).
As depicted in Figure 7, the particles typically possess sharp edges and irregular shapes, indicative of the fracture mechanism inherent to the milling process. In the MW marble, the proportion of medium-sized particles (∼60–70 μm) is notably greater relative to UW marble, while UW marble exhibits a higher volume of large-sized particles (∼500–700 μm). This finding aligns with the grain size results obtained through mineralogical analyses (Figure 2). The predominance of medium-sized grains in the MW marble contributed to a more uniform and less aggressive abrasive action against the tool surface. In contrast, the greater abundance of large grains in UW marble resulted in more concentrated impacts, leading to accelerated and uneven wear of the diamond grains. Therefore, the grain size distribution of the marble plays a pivotal role in determining the level and nature of tool wear experienced during the milling process.
The wear mechanisms observed on vacuum-brazed diamond tools—edge deterioration, fracture, and particle pull-out—directly impact the tool's mechanical integrity and service life. While the prevailing consensus in the literature suggests that harder rocks typically induce greater tool wear, 20 the findings of this study present a counter-intuitive outcome: the softer MW marble resulted in a higher tool wear rate (54%) compared to the harder UW marble (47%). This discrepancy can be attributed to the specific chip morphology of the MW marble. The softer structure of MW generated a substantial volume of ‘abrasive sludge’ within the cutting zone, where fine particles continuously eroded the tool matrix. In contrast, the chips formed during the machining of UW were larger and more readily evacuated from the cutting zone, thereby minimizing contact with the matrix. This phenomenon is significantly influenced by the petrographic characteristics observed in Figure 2. The MW sample possesses a coarser grain structure and lower mechanical strength. While this structure facilitates crack propagation along cleavage planes—resulting in a higher Material Removal Rate (MRR)—it concurrently leads to the detachment of larger, abrasive calcite macro-particles during milling. These detached particles subject the tool matrix to severe abrasive wear, explaining the elevated wear rate observed for MW. Conversely, the UW sample, characterized by a finer grain size and compact interlocking texture, exhibits higher fracture toughness. This microstructure promotes a controlled cutting mechanism rather than brittle fracture, generating finer chips and reducing the abrasive impact on the tool surface, which correlates with the lower wear rate (47%). Consequently, macro-mechanical properties, such as compressive strength, are insufficient on their own to predict the wear behavior of vacuum-brazed tools.
The intensified wear observed in MW marble can be attributed to the generation of finer abrasive debris during cutting, which increased the frequency of interaction at the grain boundaries. In contrast, the formation of larger chip grains in UW marble minimized the contact area between the abrasive particles and the tool matrix, effectively reducing material loss. These findings extend the current understanding in the literature20,23,34 by demonstrating that chip morphology and microstructural characteristics can exert a more dominant influence on wear rates than bulk mechanical hardness.
Furthermore, the simultaneous increase in Ra and SE suggests that progressive wear leads to both micro-damage on the workpiece and higher power demand. Notably, Fr remained relatively consistent throughout the process, indicating that force magnitude alone may not be sufficient to detect the onset of tool degradation. This suggests that SE and Ra serve as more sensitive indicators for monitoring tool condition than cutting force.
Consequently, minimizing tool wear is essential for ensuring product quality and operational efficiency. The results imply that process optimization and tool selection should consider chip formation behavior and petrographic texture alongside conventional mechanical properties. This approach provides a practical framework for enhancing tool longevity and implementing sustainable machining strategies in the natural stone industry.
The key results obtained from the experimental studies on the machinability of MW and UW marbles are summarized below:
Machining parameters have a distinct influence on the general machinability of marble and the wear behavior of diamond milling tools. The correct selection of parameters is critical for improving surface quality and extending tool life. As a result of wear tests, it was observed that the dominant damage types were edge and corner wear, along with the fracturing and fragmentation of diamond grains. Complete detachment of diamond grains was not observed throughout the 7 m2 machining process. A distinct difference was observed at the onset of wear; while significant fragmentation occurred after only 1 m2 of machining for MW marble, this situation started after 5 m2 for UW marble. For MW marble, surface roughness (Ra) and specific energy (SE) values showed a consistent increase as machining progressed, and this increase became pronounced especially after 3 m2; this situation is directly related to the onset of wear and grain fragmentation. In contrast, the cutting force (Fr) remained constant throughout the process, indicating mechanical stability. It was measured as 54% for MW marble and 47% for UW marble. This difference is attributed to microstructural properties, chip sizes, and the lower mechanical strength of MW marble, which led to increased material loss and higher tool wear. Overall, the microstructural properties of marble materials and the selection of appropriate machining parameters are key factors for managing tool wear and ensuring efficient processes in the marble production sector.
Future research will focus on the machinability of limestones with heterogeneous structures. Specifically, to more deeply understand the effect of mineral hardness variations on tool life, the influence of structural parameters such as porosity, grain size distribution, and fossil content on machinability indicators and tool wear mechanisms will be investigated; and a comprehensive machinability database for natural stones will be created.
Footnotes
Declaration of conflicting interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Afyon Kocatepe Üniversitesi, (grant number 23.FEN.BİL.06).
