Assessment of Sustainable and Machinable Performance Metrics of Monocrystalline Silicon Carbide Wafer with Electrophoretic Assisted Multi-Diamond Wire Sawing

The rapacious demand for energy in semiconductor wafer manufacturing industries has significant implications for global warming and wafer manufacturing costs. Assessing sustainability in the multi-diamond wire sawing (MDWS) process is crucial for reducing costs and mitigating environmental impacts. However, sustainability assessment integrated with machinability performance metrics in this process has not been investigated. This novel study extensively analyzes sustainability metrics such as processing time, energy consumption, carbon dioxide emission, machining cost, and machinability characteristics, including surface roughness, diamond wear rate, and sawing temperature in monocrystalline silicon carbide (mono-SiC) sawing process. Experiments were conducted using traditional MDWS (T-MDWS), reactive MDWS (R-MDWS), and electrophoretic-assisted reactive MDWS (ER-MDWS) coolants. An autoregressive integrated moving average (ARIMA) model were used to predict the overall energy consumption of the MDWS machine. Results showed significant improvements across various responses such as processing time, energy consumption, carbon dioxide emissions, machining cost, surface roughness, diamond wear rate, and sawing temperature, with reductions of 2.95%, 3.87%, 6.80%, 12.82%, 4.68%, 16.32%, and 4.39%, respectively. Furthermore, the ARIMA model results indicate that the total energy consumption prediction accuracy reaches 98.813%. The findings demonstrated that the ER-MDWS cooling strategy is well-suited for large-scale wafer production without compromising surface quality while minimizing environmental impact.


Introduction
Monocrystalline silicon carbide (mono-SiC) is a third-generation semiconductor material that stands out as a noteworthy polytype semiconductor material characterized by its wide bandgap, high breakdown electric field, and excellent thermal and electrical properties [1][2][3].The material has witnessed a surge in the demand for SiC wafers across diverse industries, thus prompting the need for a cost-effective and sustainable wafer fabrication process [4].
To meet this growing demand, multi-diamond wire sawing (MDWS) technology has gained widespread adoption for slicing hard and brittle materials into wafers, including mono-SiC [5].MDWS is a promising technology for slicing large-diameter hard and brittle materials, offering higher cutting efficiency and improved surface quality compared to the free abrasive wire sawing process.It is well-suited for mass production of thinner wafers in a single process [6].
However, the inherent hardness and brittleness nature of mono-SiC, combined with the demand for thinner wafers, result in a reduced material removal rate (MRR) and prolonged processing/machining times.Consequently, this leads to increased energy consumption and carbon dioxide emissions from MDWS machines.These factors emphasize the demand for a sustainable SiC slicing process to reduce environmental footprints.Sustainable manufacturing is the practice of producing goods in an environmentally responsible manner, with an emphasis on minimizing resource consumption, reducing waste, and promoting social and economic well-being throughout the production process [7][8][9].To attain a sustainable machining process, it is crucial to create eco-friendly conditions and efficiently manage the energy consumption and carbon dioxide emissions of the machine.A key aspect of sustainable machining revolves around the use of sustainable coolants, materials, and technologies which have a notable influence on the machining characteristics and sustainable performance metrics [10][11][12][13].However, studies endeavors pertaining to analysis of sustainability performance metrics in the DWS process are limited, and a few studies have investigated the machinability performance metrics such as surface quality and MRR by hybridizing the DWS process: electrical discharge-assisted DWS [14], electrochemical discharge-assisted DWS [15], and ultrasonic-assisted DWS process [16] are discussed below.Yan, et al. [17] developed a novel 3D ultrasonic vibration-assisted diamond wire sawing process along with a corresponding mathematical model to enhance the machining characteristics of SiC.The findings demonstrate that the ultrasonic vibration-assisted system efficiently mitigates sawing forces compared to conventional processes.Gao, et al. [5] investigated the subsurface damage depth of the SiC sawing process using numerical and finite element approaches.Results show that minimal subsurface damage of 9.46μm is discerned at a sawing depth of 5 mm, with cutting parameters such as wire speed and feed rate exerting notable influence on the extent of subsurface damage.Wang, et al. [18] investigated the effect of wire speed on subsurface cracks in mono-SiC using governing equation.Their findings indicate that wire speeds below 100 m/s imparted minimal subsurface cracks.Huang, et al. [19] studied the machining characteristics and surface profile of mono-SiC during the DWS process, and results show that a consistent surface roughness ranging from 0.6 to 0.8 μm has been observed in all experiments.The author affirms the absence of a discernible relationship between the surface roughness and the utilized process parameters.Furthermore, it is confirmed that the specific cutting energy exhibits an inverse proportionality to the MRR.Hardin, et al. [20] investigated the surface and subsurface damage of SiC under the rocking mode, revealing that a reduced diamond abrasive size has a notable positive impact on the wafer's surface and subsurface damage.Additionally, the author underscores the significant influence of the feed rate on the abrasive diamond grit wear and overall surface quality.Cvetković, et al. [21] examined the sawing performance of monocrystalline and polycrystalline SiC employing ultra-precision dicing and wire sawing processes.Results show that wire sawing proves to be a superior approach for attaining precise surface roughness in SiC wafers.Wang, et al. [22] investigated the cutting performance of the DWS process for SiC wafer processing.Results indicate that when the diamond wire diameter and wire speed increase, it leads to an improved MRR and decreases the diamond wear rate; however, the surface roughness of the wafer and total thickness variation increase.Wang, et al. [4] proposed a theoretical nano-and micro-scratch force model based on the actual diamond abrasive shape for enhancing the surface quality of mono-SiC as-sawn wafers.
The finding indicates that the results of the developed theoretical model are close to the experimental results and that the model is beneficial for improving the surface quality of the as-sawn wafer.
According to the current state-of-the-art statistics, previous studies have primarily focused on critical aspects such as diamond wear rate, surface quality, sawing force, and MRR of various hard and brittle materials.However, sustainable performance metrics of the MDWS process have not been investigated.Consequently, this study addresses the gap by assessing the sustainability performance metrics, including energy consumption, carbon dioxide emission, processing time, machining cost, and machinability characteristics such as surface roughness, wire wear, and sawing temperature.
Three distinct cooling strategies such as traditional MDWS (T-MDWS), reactive MDWS (R-MDWS), and electrophoretic-assisted reactive MDWS (ER-MDWS) coolants, are proposed to evaluate the sustainability and machinability performance metrics of the sawing process.The findings indicate that ER-MDWS and R-MDWS have better advantages than T-MDWS in terms of sustainability and machinability performance metrics.

Materials
The sawing experiments were performed using a WS-150 MDWS machine (SETEC Corporation, Taiwan) under both traditional and sustainable sawing processes, as illustrated in Fig. 1.Cylindrical N-type mono-4H-SiC (Winsheng Material Technology, Taiwan) has been used as a work material and sliced into a 200 μm thickness employing Nibased electroplated multi-diamond wire (Asahi Diamond, Japan) with a total diameter of 150 μm and an average abrasive diamond size of 11.77 μm.Controllable process parameters such as wire speed, feed rate, and wire tension were utilized, and their experimental design details are shown in Table 1.Three different cooling conditions, as depicted in Table 2, traditional MDWS (T-MDWS), reactive MDWS (R-MDWS), and electrophoretic-assisted reactive MDWS (ER-MDWS), were employed to enhance the sustainability and machinability performance metrics of the sawing processing.The energy consumption of the machine has been measured using an industrial control products data acquisition system power meter, a PM-3133 model (ICP DAS, Taiwan).The measured data has been controlled using a data acquisition system (PMC-5141) with different power sources: a programmable logic controller (PLC), cooling, wire drivers, servo drivers, and a table feed system.An Optris CS-LT infrared pyrometer (Optris, Germany) has been used to monitor the transient sawing temperature at different cutting depths.The temperature data has been directly transferred to a personal computer via a USB programming kit and filtered using LABVIEW software.The schematic diagram of electrophoretic deposition is illustrated in Fig. 2.

Mechanism of Electrophoretic Assisted-Multi-Diamond Wire Sawing
Electrophoretic deposition involves the movement of charged particles in a stable colloidal suspension through a liquid under the influence of an electric field.These particles are then deposited onto a substrate with an opposite charge, resulting in the formation of the desired material [23,24].This study employs a novel hybrid technique, electrophoretic deposition of reactive coolant in the multi-diamond wire sawing (ER-MDWS) to slice 4H-SiC as illustrated in Fig. 2.This method involves the deposition of a reactive coolant onto the sawing region through an electrophoretic deposition cooling system, aiming to enhance sawing efficiency and reduce the environmental footprint.The process involves using Fenton powder (Fe3O4) as a reactive abrasive, which is brought close to the ingot surface to initiate the conversion of hydrogen peroxide (H2O2) into hydroxyl radicals through the Fenton reaction.
These hydroxyl radicals effectively break the C-Si bonds on the SiC surface, leading to the formation of an oxide layer [24][25][26], as described in Eq. (1) up to (3).Furthermore, Fe 3+ ions transform into Fe 2+ ions, actively participating in the Fenton reaction, as shown in Eq. ( 4) and ( 5).This study modified the reactive coolant with a cationic surfactant to induce selective electrophoretic deposition (SEPD).This modification increased, thereby elevating the zeta potential of the Fe3O4 particles intended for coating onto the wire used in processing the ingot surface.

Measurements and Analysis of Sustainable Performance Metrics
Assessment of sustainable performance metrics in the manufacturing process is crucial for evaluating and enhancing production's environmental, social, and economic aspects [7,27,28].This study considers four sustainable performance metrics to assess the sustainability of the MDWS process, including processing time, energy consumption, carbon dioxide emission, and machining cost.
Processing Time.Processing time that incorporates the machine's standby time (T Sd ), cooling time (T Cool ) and sawing time (T Sw ).

Energy Consumption. The energy consumption of the MDWS machine is classified as variable and constant
depending on the factors that affect the machine units and is measured by a digital power meter.

Carbon Dioxide Emission.
The CO2 emissions resulting from electrical energy, considering both constant and variable energy consumption of the machine, and the CO2 emissions originating from materials including coolant, work material, and diamond wire, have been analyzed based on the GHG emission protocol.
Machining Cost.Machining costs have been analyzed, encompassing electricity, depreciation, labor, machine tools, and cooling system costs.

Measurements of Machinable Performance Metrics
Machinability performance measurements involve assessing or evaluating how effectively a material or process can be machined.This study considers three machinable performance characteristics: surface roughness, wire wear, and sawing temperature.
Surface Roughness.The as-sawn wafer's surface roughness (Ra) has been measured five times at uniform intervals, considering the wire entrance, midpoint, and exit points along the feed direction using Taylor Hobson coherence correlation interferometry (CCI, Taylor Hobson, Leicester, UK).Moreover, the subsurface damage of the wafer has been examined using a focused ion beam (Quanta 3D FEG, FEI, Czech Republic) machine at three different points and averaged.

Diamond Wear Rate.
A virgin wire has been employed to assess the diamond wear rate under different cooling conditions, and its wear rate has been analyzed by observing the diamond wire micrographs using a Phenom XL scanning electron microscope (SEM, Thermo Fisher Scientific, Netherlands) and tensile strength before and after the sawing process.
Sawing Temperature.The transient sawing temperature has been measured using a CS-LT infrared pyrometer (Optris, Germany) at different contact lengths of the wafer within a response time of 10 ms/data.The collected data has been filtered using LABVIEW software.

Energy Consumption Prediction
Prediction of the energy consumption of the machine is carried out through a time series algorithm known as the autoregressive integrated moving average (ARIMA).It is a linear regression model specifically designed for forecasting time-series data.It is denoted as ARIMA (p, d, q), with 'p' representing the auto-regressive part's order, 'd' indicating non-seasonal variation, and 'q' specifying the order of the moving average part [29,30].The model has been built using Eq. ( 6) Where  is the lag operator,   represents the autoregressive parameter,   indicate the moving average parameters,  is the error term, and  is the intercept.In this study, the ARIMA (p, d, q) model has been used in the order of (1, 1, 1).

Results and Discussions
This study analyzed the experimental outcomes while slicing 4H-SiC wafers using conventional and sustainable cooling environments.A comprehensive sustainability and machinability performance metrics results obtained from various cooling conditions are presented in Error!Reference source not found.. Table 3. Experimental results of sustainability and machinability performance metrics.

Processing Time
An efficient manufacturing process aims to minimize processing/machining time without compromising the surface quality of the product [31].The input process parameters and cutting conditions influence the processing time, and to minimize the total processing time, various cooling strategies have been employed to incorporate the machine's standby time (T Sd ), cooling time (T Cool ) and sawing time (T Sw ).Eq. (7) gives the total processing time resulting from each cooling environment.  =   +   +   (7) where   is the total processing time for the MDWS process (ℎ),   is the total standby time (ℎ),   is cooling time (hr.) and   is the sawing time (hr).Fig. 3 illustrates the processing time in various cooling conditions.
Consequently, a minimal processing time of 24.01 hrs.has been observed for ER-MDWS, followed by 24.Moreover, as the diamond wire lifespan extends, there is a decrease in wire breakage occurrences and time required for replacement.Consequently, it can be concluded that reducing the processing time is an effective strategy for minimizing sustainable performance metrics.

Energy Consumption
Precise assessment of the energy consumption of a machine can be achieved by considering value-added and nonvalue-added operations during the machining process [32].In this study, the energy consumption of the MDWS machine is categorized as variable (E variable ), and constant (E Constant ) energy consumption, and both have been considered value-added operations.The total energy consumption (E total ) has been calculated using Eq.(8) up to (10).
=   +    +   +    +    +    (10) where   is the total energy consumption of the machine (kW.h),   is the energy consumption of standby (kW.h),    is the energy consumption of the PLC controller (kW.h),   is the energy consumption of the coolant pump (kW.h),    is the energy consumption of the wire drive system (kW.h),   is the energy consumption of the feed drive system (kW.h), and    is represents the energy consumption of the servo driver system (kW.h).The findings show that the total energy consumption has been lowest under ER-MDWS

Carbon Dioxide Emissions
The manufacturing industries contribute significantly to carbon dioxide emissions, primarily due to energy consumption, raw material extraction, transportation, and industrial processes [33,34].The CO2 emissions resulting from the energy consumption of the machine and materials are analyzed using Eq. ( 11).The results are depicted graphically using a Sankey diagram, as illustrated in Fig. 7, and the CO2 emission factors of materials are listed in Where   is the carbon dioxide emission from materials and   is denoted the carbon dioxide emission from the energy consumption of the machine.Carbon Dioxide Emission from Materials.A significant amount of carbon is being released into the atmosphere as a direct consequence of the materials used during the manufacturing process.This elevated carbon dioxide emission level is a notable environmental concern, contributing to the overall impact on the planet's carbon footprints [37,38].In this study, the CO2 emissions from the work material, coolant, and diamond wire have been computed using Eq. ( 12), and the cumulative CO2 emissions from these materials are presented in Fig. 6.
where   is the CO2 emissions of the work materials (kgCO2e),   is the volume of the sliced wafers (m 3 ),   is the density of the material (kg/m 3 ), and   is the CO2 emission factor of the material (kgCO2e),   denotes the carbon dioxide emissions of the coolant (kgCO2e),   denotes the slicing length of the k th part (mm),   denotes the slicing thickness of the i th part (mm),   denotes the volume of the concentrated coolant (L),   denotes the carbon dioxide emission factor of the mineral oil (kgCO2e),   denotes the carbon dioxide emissions factor of the water (kgCO2e) and   is the sawing area of the i th part, C DW denotes the CO2 emissions from the diamond wire during the slicing of a single wafer (kgCO2e), δ k is the slicing thickness of the k th part (mm), L W is the length of the diamond wire (m), A W is the mass per unit length of the diamond wire (g/m), C W is the CO2 emissions of the diamond wire (kg), nw denotes the number of sliced wafers in a single process, and the other signs are as mentioned previously.The life cycle inventory of the Fenton reaction with different chemicals constituting the reactive coolant, combined with the electricity consumption during preparation, has been analyzed (see Table A.1).The finding indicates that the highest CO2 emission has been observed in the case of the ER-MDWS (1.55 kg/CO2e), followed by the R-MDWS (1.12 kg/CO2e) and T-MDWS (1.03 kg/CO2e) because of the increase in the number of participating cooling materials constituting the reactive coolants and its carbon dioxide emission factor.When the number of participating cooling materials and their respective ratios increased, the aggregated CO2 emissions from the materials increased.However, the ER-MDWS significantly improves the MRR, leading to minimizing carbon dioxide emissions from the machine.
Consequently, to minimize CO2 emissions from materials, T-MDWS is an effective cooling condition compared with ER-MDWS and R-MDWS.Carbon Dioxide Emissions from Energy Consumption.The CO2 emissions resulting from both the constant and variable electric energy consumption required for the sawing process have been computed using Eq. ( 13) up to (16).
Where   and   is the carbon dioxide emissions from constant and variable energy consumption of the machine, respectively,   denotes the carbon dioxide emissions of the electric energy consumption of the PLC control systems,   denotes the input power of the PLC control system (kW.h),  denotes the input power of the cooling system,  denotes the operating time, and α represents the carbon dioxide emissions coefficient of the country's electricity.A CO2 emission factor of 0.7773 kgCO2e/kW.hhas been taken from the national average emission factor of power grids in Taiwan.Results showed that the CO2 emission from T-MDWS, R-MDWS, and ER-MDWS varied from 285.292 kgCO2e, 276.14 kgCO2e, and 265.29 kgCO2e, respectively.Among these, ER-MDWS yielded the lowest CO2 emissions due to its minimum processing time, while the highest CO2 emission has been observed in the T-MDWS process.Reactive coolants can smooth the top surface of SiC, reducing processing time and minimizing energy consumption and CO2 emissions.Consequently, the total CO2 emissions of the T-MDWS, R-MDWS, and ER-MDWS have been analyzed using Eq. ( 11), resulting in values of 286.322 kgCO2e, 277.26 kgCO2e, and 266.84 kgCO2e, respectively as shown in Fig. 7. Notably, the utilization of electrophoretic-assisted reactive coolant significantly enhanced the sustainability performance metrics of the response of the study.However, it is essential to note that a cooling strategy alone is not the sole alternative for achieving a significant change; it requires optimal sawing parameters to be genuinely effective.

Machining Cost Analysis
In sustainable and advanced manufacturing processes, cost-effectiveness and resource-efficient manufacturing processes are a priority for achieving long-term economic viability and reducing the environmental footprint [39][40][41].
This study performed a comparative machining cost analysis in different cooling conditions, as presented in Table 5.

Solar energy xx (2021) xxx-xxx
process is assisted by a reactive coolant, which not only reduces processing time but also minimizes the energy consumption of the machine.Also, the overhead cost, calculated by considering the cost of operator and depreciation, reveals that the minimum overhead costs are observed in Fig. 8(b) ER-MDWS and R-MDWS, respectively, due to its total processing time.Consecutively, the cost of the diamond wire tool and each coolant has been calculated based on the effective length of the diamond wire and the number of elements used for preparing the coolants.Results are shown in Fig. 8(c, d), where the minimum diamond wire cost has been observed in ER-MDWS, R-MDWS, and T-MDWS, respectively.Moreover, the minimum coolant costs are observed in the case of T-MDWS, R-MDWS, and ER-MDWS, respectively, due to the number of elements used in the coolant.The overall machining costs are reduced by 12.82% and 5.21% using the ER-MDWS and R-MDWS sawing approaches, respectively, compared to the T-MDWS, as illustrated in Fig. 8(e).The lower cost is associated with the ER-MDWS because of its higher MRR and minimum processing time.Electrophoretic deposition and reactive coolant played pivotal roles in machining cost; however, it should be noted that the cost of the cooling condition in ER-MDWS has been higher than that of T-MDWS.
Table 5. Overall costs at different cooling conditions.

Surface Roughness
In the MDWS process, surface quality is a crucial indicator for assessing its quality and process efficiency.This, in turn, directly impacts the subsequent processing costs [42,43].SiC has covalent Si and C bonds with two different polarities of Si and C-face, leading to notable variations in its physical, chemical, and electrical properties [44].These phenomena significantly influence the surface quality of the SiC wafers.In this study, the surface roughness of the Furthermore, the subsurface damage of the wafer has been measured using a focus ion beam (FIB) machine at three points.The minimal average subsurface damage of 0.8 μm and 0.61 μm has been observed at the Si and C-face of SiC in the ER-MDWS process illustrated in Fig. 9(d).Therefore, it can be asserted that a diamond wire coated with a reactive abrasive has the potential to minimize the subsurface crack depth in the as-sawn wafer.

Diamond Wear Rate
The wire wear rate of the diamond plays a significant role in the surface quality of as-sawn wafers, and it quantifies the rate at which the abrasive diamond particles wear during the cutting process [47].In the wafer sawing process, the machinability characteristics, encompassing surface roughness, subsurface damage, and sawing temperature, are contingent upon and directly affected by tool wear.It impacts various parameters, including wire speed, feed rate, cutting depth, and cooling conditions [42,48,49].In this study, the feed rate and wire speed were set at 0.1 mm/min and 10 m/sec to assess the influence of various coolants and its lubricating strategies on diamond wear rate.In each experiment, a virgin diamond wire with particular coolants have been used, and the SEM result, as illustrated in Fig. compared with the new diamond wire.Therefore, it can be asserted that uniform deposition of a reactive coolant onto the diamond wire significantly contributes to minimizing the diamond wear rate.The SEM morphology and tensile strength results show that ER-MDWS is a promising cooling method for minimizing the wire wear rate compared with R-MDWS and T-MDWS.Moreover, the rocking mode also has a significant impact on minimizing the wire wear rate owing to the minimum contact length between the diamond wire and work material.

Sawing Temperature
The input parameters, including wire speed and feed rate, and the contact length between the wire and work material directly influence the sawing temperature, which is the primary reason for generating residual stress and undesired distortions in the sawn wafers [50].In addition, the cooling phenomenon significantly affected the generated sawing temperature.This study employed different cooling conditions and rocking-mode sawing processes as alternative solutions to minimize the sawing temperatures.Results show in Fig. 11 that the maximum sawing temperature of 36

Holistic Assessment
This study assesses the sustainability and machinability performance metrics under T-MDWS, R-MDWS, and ER-MDWS conditions, utilizing a radar chart depicted in Fig. 12.The scores for each metric are adjusted to a range of 0-30, allowing for a comprehensive evaluation of the influence of each sawing conditions.As a comprehensive evaluation of swing performance metrics, the ER-MDWS is an excellent cooling medium for improving the sustainability and machinability performance of the MDWS process.Overall, excellent results have been obtained with the ER-MDWS, followed by the R-MDWS.It can be noted that the Fenton reaction, uniformly deposited onto the diamond wire through the aid of an electric field, significantly smoothing the top surface of SiC by effectively breaking the C-Si bonds on the SiC surface with hydroxyl radicals.This enhancement in the MRR leads to a reduction in processing time and energy consumption.Hence, it is noteworthy to mention that ER-MDWS demonstrates superior sustainability and machinability performance when compared to R-MDWS and T-MDWS.

Conclusion
This study provides a comprehensive sustainability assessment, including processing time, energy consumption, carbon dioxide emission, machining cost, and machinability characteristics such as surface roughness, diamond wear rate, and sawing temperature of mono-SiC slicing under various cooling strategies.The key findings of this study can be drawn as follows: • In the case of ER-MDWS and R-MDWS, the total energy consumption is decreased by 3.87% and 2.127%, respectively, compared to the T-MDWS process.This reduction in energy consumption can be attributed to the reactive coolant coverage of the diamond abrasives in ER-MDWS and R-MDWS, which helps maintain the sawing edges, minimizes sawing forces, and facilitates efficient material removal.Furthermore, the ARIMA prediction model showcased a high accuracy in predicting total energy consumption, achieving a rate of 98.813%, closely aligning with the experimental results.
• In the ER-MDWS and R-MDWS conditions, total carbon dioxide emissions of 6.8% and 3.16% were decreased, respectively.Adopting a reactive coolant plays a crucial role in minimizing carbon dioxide emissions associated with the energy consumption of the MDWS machine.Notably, the T-MDWS exhibits the lowest carbon dioxide emission from materials, followed by R-MDWS and ER-MDWS, which is attributable to the composition of the cooling materials.
• A reduction in total sawing cost has been observed, with decreases of 12.82% and 5.21% in the ER-MDWS and R-MDWS processes, respectively, compared to T-MDWS.
• The surface roughness of (Ra) decreased by 4.68% and 3.12% under ER-MDWS and R-MDWS, respectively, compared to the T-MDWS cooling condition.Furthermore, minimal subsurface damage of 0.6μm has been observed at the C-face of the sawn wafer during ER-MDWS.In ER-MDWS, the MRR is facilitated by reactive chemical coolants, reducing impact forces between the diamond abrasive and work material and smoothing the top layer of the SiC surface.
• A substantial reduction in the diamond wear rate has been achieved through ER-MDWS, showing a decrease of 12.02%, followed by R-MDWS with 15.43% and T-MDWS with 26.38% compared with the new diamond wire in terms of tensile strength.This reduction can be attributed to the interaction of the Fenton reaction with SiC, enhancing the MRR and minimizing wear on the diamond, thereby contributing to the observed decrease in wear rate.
• In comparison ER-MDWS and R-MDWS with the T-MDWS cooling exhibited a maximum sawing temperature of 36.4 ˚C, attributed to the increased interaction between the diamond wire and the work material.
• Notably, the ER-MDWS cooling condition played a pivotal role in significantly improving the sustainability and machinability performance metrics of the MDWS process.Furthermore, this condition yielded the minimum diamond wear rate and surface roughness results.There exists a potential to explore a trade-off relationship between surface roughness, diamond wire life, and energy consumption by potentially increasing the feed rate under the ER-MDWS cooling conditions.
48 hrs.for R-MDWS and 24.74 hrs.for T-MDWS, with constant cutting parameters.When comparing processing times across various cooling conditions, ER-MDWS showed a reduction of approximately 2.95% compared to T-MDWS.The decrease in processing time is associated with sustainable cooling conditions (ER-MDWS) and prolonged diamond wire lifespan, taking into account the cumulative standby time, cooling time, and sawing time.The Fenton reaction contributes to the smoothing of the SiC top surface, enhancing MRR and resulting in a reduction in processing time.

Fig. 3 .
Fig. 3. Total processing time under various cooling environments

( 27 .
06 kW.h), followed by R-MDWS(27.55kW.h) and T-MDWS (28.15 kW.h), while maintaining the same sawing parameters.The Fenton reactive coolant induced a significant softening effect on the top surface of SiC, attributed to the hydroxyl radical breaking the C-Si bonds on the SiC surface.This resulted in an enhanced MRR and reduced sawing drag force.Additionally, using an electrophoretic deposition strategy can enhance the uniformity of the deposited thickness, and abrasive particles are easily migrated or deposited on the diamond wire.Due to this, the energy consumption of ER-MDWS is 3.86% lower than the R-MDWS conditions.Fig.4(a) shows that the individual Eyob Messele et al.Solar energy xx (2021) xxx-xxx energy consumption of the machine and the maximum energy consumption have been observed at the variable energyconsuming machine units.However, holistic minimum energy consumption has been observed in both ER-MDWS and R-MDWS compared to T-MDW, as shown in Fig. 4(b).It is worth mentioning that the utilization of electrophoretic deposition and reactive coolant minimized the energy consumption of the machine as well as processing time.According to observations of individual energy consumption of the machine units, utilization of an electrophoretic cooling system affects the processing time, resulting in decreased energy demand of the machine.This study confirmed that cooling strategies significantly influence the processing time, energy consumption of the machine, carbon dioxide emissions, and machining cost.Moreover, the energy consumption of the machine has been predicted by ARIMA model using Google Colab with Python (version 3.7.13).The predicted results demonstrated impressive accuracy, with each machine unit's energy consumption prediction exceeding 93.22% and an overall model accuracy surpassing 98.813% for the total predicted energy consumption.The predicted energy consumption of the machine is in good agreement with the actual measurements, as shown in Fig. 5.

Fig. 4 .
Fig. 4. Energy consumption demands: (a) individual machine units and (b) total energy consumption of the machine.
The machining costs encompass the electric consumption, diamond wire, cooling material, overhead costs, and total costs under T-MDWS, R-MDWS, and ER-MDWS conditions.The electric power consumption cost has been analyzed by considering the power consumption of the MDWS machine for each experiment and processing time.Results shown in Fig.8(a) indicate that the minimum energy consumption of the machine has been observed in the case of the ER-MDWS process, followed by the R-MDWS and T-MDWS processes.It is worth mentioning that the sawing Eyob Messele et al.

Eyob
Messele et al.Solar energy xx (2021) xxx-xxxsawn wafer has been examined using a coherent correlation interferometer (CCI), and the results shown in Fig.9(a)visually represent the wafer topography, illustrating how the surface roughness (Ra) is affected by the cooling conditions.Measurements have been conducted on both Si and C-faces, and results are illustrated in Fig.9(b, c), indicating that the surface roughness values of Ra ≤ 1 μm and Rz ≤ 8 μm.According to modern abrasive machining technology[45,46], the obtained surface roughness results are considered to have a higher surface quality.At the Cface of the sawn wafer, ER-MDWS and R-MDWS demonstrated lower Ra values of 0.122μm (4.68%) and 0.124μm (3.12%), respectively, compared to T-MDWS.Compared with T-MDWS, reactive coolant, and electrophoretic deposition increase the MRR by softening the top surface layer of SiC and reducing wafer-edge breakage, improving surface roughness.As the reactivity of the coolant increased, the MRR also increased, resulting in minimum surface roughness.However, in the case of the T-MDWS, the MRR is solely dependent on the interaction between the diamond wire and the work material.This can contribute to increased surface roughness and wire wear rate.
10(a-d), shows that the new abrasive diamond size decreased from 11.77μm to 4.67μm, 6.78μm and 7.22μm in the T-MDWS, R-MDWS, and ER-MDWS, respectively.The diamond abrasive became blunt and round when the wire cutting depth increased.However, in ER-MDWS and R-MDWS, a moderate wear rate has been observed due to the diamond wire being covered by the reactive coolant and the diamond abrasive being highly blunt in the case of T-MDWS.A diamond wire coated with a reactive coolant can lead to a reduction in direct impact loads on the tip of the diamond abrasive.Additionally, the SEM micrograph shows that in the case of ER-MDWS, a tremendous amount of reactive coolants are deposited on the diamond wire compared to R-MDWS.Consequently, a diamond pit has been observed in the case of T-MDWS.Hence, it is crucial to emphasize that the electrophoretic deposition system consistently and uniformly deposits the reactive coolants.Furthermore, the tensile strength of the diamond wire has been examined using a universal tensile-testing machine, and results, as shown in Fig.10(e), indicate that the tensile strength decreased by 26.38%, 15.43%, and 12.02% in the T-MDWS, R-MDWS, and ER-MDWS, respectively

Fig. 9 . 15 342Fig. 10 .
Fig. 9. Surface quality of as-sawn wafer: (a) 2D and 3D surface topography, (b and c) average surface roughness at the Si and C-face, respectively, and (d) average subsurface damage at Si and C-face.

. 4 °C followed by 35 . 5
˚C and 34.8 °C has been observed under the T-MDWS, R-MDWS, and ER-MDWS cooling conditions, respectively.When reactive cooling conditions assist the MRR with the electrophoretic-assisted reactive coolant, the charged particles migrate toward the top surface of the work material and soften, resulting in reduced friction between the diamond wire and the workpiece.This, in turn, improves the MRR and reduces the sawing temperature.Moreover, the temperature variations at different depths of cuts are shown in Fig. 11, indicating that the sawing temperature increased gradually up to the center point of the ingot and decreased toward the bottom due to the contact length between the diamond wire and the work material.The coolant temperature increased from 18 °C to 24 °C because of the coolant recycling process and convection heat transfer.Consequently, the recycling of coolants increased the sawing temperature.The absence of a reactive coolant has the most prominent effect on the increase in sawing temperature during the T-MDWS.

Fig. 12 .
Fig. 12.Comparison of sustainability and machinability performance metrics of the MDWS process.

Table 2 .
Chemical composition of coolant materials.

Table 4
and Table A.1.

Table 4 .
CO2 emission factors of materials.