Optical Emission Spectroscopy and Ultrafast Arc Management Strategy on the Quality of AlN Thin Films Using a 13.56 MHz RF Generator

Optical Emission Spectroscopy and Ultrafast Arc Management Strategy on the Quality of AlN Thin Films Using a 13.56 MHz RF Generator | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Optical Emission Spectroscopy and Ultrafast Arc Management Strategy on the Quality of AlN Thin Films Using a 13.56 MHz RF Generator Yu-Shin Chen, Cheng-Yuan Kao, Hsuan-Fan Chen, Chih-Hao Tsui, Ting-Yueh Yang, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6737015/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 24 Jul, 2025 Read the published version in The International Journal of Advanced Manufacturing Technology → Version 1 posted 5 You are reading this latest preprint version Abstract This study focuses on the obtaining quality of arc-managed aluminum nitride (AlN) films in a PVD plasma system using a 13.56 MHz RF power source. Key parameters such as reflection coefficient (Г), suppression time, and re-ignition time were studied with the aim of reducing arc interference, increasing film crystallinity, and improving surface morphology. The key innovation of this work lies in the implementation of an ultrafast arc management strategy, with suppression/re-ignition times as short as 2 µs/2 µs—significantly shorter than those typically reported in the literature (50–1000 µs). This ultrashort modulation intercepts arc events at their micro-discharge stage, effectively minimizing energy accumulation and preventing the formation of high-energy "hard arcs." Consequently, this approach stabilizes plasma conditions and leads to significant improvements in film crystallinity, surface smoothness, and deposition rate uniformity, offering a new pathway toward high-performance AlN thin film fabrication in RF sputtering systems. In the plasma spectra obtained from in situ optical emission spectroscopy (OES), significantly higher spectral intensities were observed during radio frequency (RF) sputtering, particularly the emission feature associated with ionized nitrogen (N₂⁺) at approximately 390.93 nm. This suggests that RF plasma provides energy conditions favorable for the formation of high-quality aluminum nitride (AlN) films. Furthermore, we applied the principal component analysis (PCA) algorithm for big data analysis to reduce dimensionality and visualize the clustering patterns of OES data recorded during thin film deposition under three different suppression/re-ignition conditions: 2 µs/2 µs, 100 µs/50 µs, and 1000 µs/500 µs. Reactive magnetron sputtering Optical emission spectroscopy Radio frequency (RF) 13.56MHz Aluminum nitride (AlN) Arc management Big data analysis Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 1. Introduction AlN has garnered substantial attention as a potential material for microelectronic and optoelectronic applications, including short-wavelength light emitters. Its exceptional properties, such as a wide band gap of 6.2 eV, excellent thermal conductivity, low thermal expansion coefficient, outstanding chemical and thermal stability, high dielectric strength, and fast surface acoustic wave velocity, make high-quality AlN films highly applicable in high-frequency and high-power device technologies [ 1 – 7 ]. Arcing problems are frequently encountered in semiconductor manufacturing, particularly at low pressures, high plasma power, and high plasma density. These issues are exacerbated in nanoscale devices, which are more vulnerable to damage caused by arcing. In CVD tools, common issues include wafer-level and electrostatic chucks (ESC) arcing in high density plasma (HDP) chambers, as well as wafer-level arcing in PECVD chambers. Arcing can lead to physical damage to chamber components, contamination, and RF system malfunctions. On device wafers, arcing can cause issues such as dielectric breakdown, interconnect failures (open or short circuit), contamination, and gate oxide integrity (GOI) failure. Identifying and addressing the root causes of arcing is essential to minimize reduced mean time between failures (MTBF) and low up-time, which can negatively affect overall fab efficiency [ 8 ]. Arcing has long been a known issue in DC-powered plasma systems, especially in dielectric sputtering, caused by charge buildup and electrical breakdown on dielectric films deposited on targets and chamber walls [ 9 ]. To address this, DC supplies now include advanced arc management systems that detect arcs and respond by briefly cutting power or reversing output polarity [ 10 ]. In critical processes, the downtime is factored into the total process time (joule mode). Periodic pulsing or polarity reversal can also reduce arc occurrence. RF power emerged as an alternative for direct sputtering of insulators without arcing [ 11 ], but recent studies show RF systems can still experience arcs, which may be highly disruptive. Causes include charge buildup on gate electrode patterns [ 11 ], polymer coatings [ 12 ], hardware defects, damage to anodized layers, electrical potential differences between components [ 13 ], and high RF power [ 14 ]. This manuscript presents recent findings on RF arc detection and management. In magnetron sputtering, arcing remains a major challenge, mainly causing particle generation. Traditional arc management detects high current or low voltage anomalies and temporarily interrupts power to extinguish arcs [ 15 ]. Although improvements have reduced arc energy, current systems cannot react fast enough to fully prevent particle formation. The generated particles are often much larger than the film thickness, leading to serious defects and quality issues. While arc management reduces particle quantity, it remains a reactive strategy that cannot prevent particle ejection at the arc’s onset [ 16 ]. Arcing is also a major cause of process control and thin film quality issues in magnetron sputtering. Unlike the stable glow of sputtering plasmas, arcs involve localized, intense discharges caused by collective electron emission during breakdown events [ 17 ]. The concentrated energy can produce molten material and eject macroparticles several microns in size, regardless of material or gas type. These macroparticles cause surface defects, disrupt power delivery, lower deposition rates, and result in non-uniform thickness and banding. Arcs often originate from localized charge buildup on particles, impurities, or flakes on the target surface under high ion flux conditions [ 18 ]. In this study, we investigate for the first time the impact of ultrafast suppression/re-ignition times on film quality, filling a gap in the existing literature. The reason why using ultrafast suppression/re-ignition time as small as 2 µs lies in its ability to promptly terminate the early-stage arc formation before it evolves into a fully developed micro-arc. This rapid interruption effectively reduces localized thermal accumulation, prevents excessive energy concentration, and facilitates a more uniform energy distribution across the deposition surface. As a result, the plasma remains in a more stable and controlled regime, promoting improved film uniformity, reduced surface roughness, and enhanced overall quality. While previous studies have predominantly investigated suppression/re-ignition intervals of 10 µs [ 19 ] and 50 µs [ 22 ], the potential of employing an ultra-short 2 µs interval has remained largely unexplored in the context of micro-arc discharge control. In this work, our experimental results demonstrate that the implementation of a 2 µs suppression/re-ignition strategy significantly improves film quality, indicating its superior effectiveness in mitigating micro-arc phenomena. The key advantage of ultrafast suppression/re-ignition lies in its ability to quickly respond to micro-arcs, reducing arc-induced defects and minimizing the impact on deposition rate. Effective arc management requires both fast detection and immediate power cutoff to limit arc energy. When combined with different reflection thresholds, this method enables more precise control. Overall, this approach offers a practical and theoretical advancement over conventional arc control techniques for thin film deposition. Optical emission spectroscopy (OES) is a non-invasive diagnostic tool for plasma analysis, capable of detecting light signals emitted by excited ions, atoms, and molecules within the plasma.[ 42 , 43 ] By comparing these spectral lines with reference databases, the corresponding chemical species can be identified. This technique has been widely applied in semiconductor processes such as etching, PECVD, and PVD.[ 44 – 46 ] Moreover, previous studies have indicated a correlation between spectral signals and thin film quality characteristics.[ 45 , 47 ] However, OES typically generates large volumes of data, making it challenging to extract meaningful information for classification purposes. Therefore, in this study, principal component analysis (PCA) was employed to reduce the dimensionality of OES data by transforming it into a few principal components, thereby capturing the most essential features and exploring potential correlations with film quality and different arc management strategies. 2. Arc detection and arc management Arc detection is based on monitoring the reflection coefficient (Γ), which quantifies the ratio of reflected to forward power. When an arc occurs, abrupt plasma impedance changes cause a significant spike in Γ. The RF generator’s suppression mechanism is triggered once Γ exceeds a predefined threshold, allowing real-time adaptation to varying process conditions. Arc detection in the RF generator is triggered using the reflection coefficient, which ranges from 0 to 1. When an arc occurs, it causes a sudden change in plasma impedance, leading to a significant increase in the reflection coefficient. The RF generator continuously monitors this value, and if it exceeds a predefined adjustable threshold within the 0 to 1 range, the system identifies an arc and activates the suppression mechanism. This adjustable threshold allows for fine-tuning to accommodate different process requirements, ensuring precise and efficient arc detection and management [ 19 ]. $$\:\varGamma\:=\sqrt{\frac{{P}_{re}}{{P}_{for}}}=\frac{{z}_{L}-{z}_{0}}{{z}_{L}+{z}_{0}}=\left|\frac{1-VSWR}{1+VSWR}\right|$$ 1 Based on Eq. ( 1 ) for voltage standing wave ratio (VSWR), the reflection coefficient (Γ), calculated as the ratio of reflected power (P re ) to forward power (P for ) or derived from the load (Z L ) and characteristic (Z 0 ) impedance serves as a key indicator for arc detection, with an adjustable threshold 0–1 to trigger suppression mechanisms when plasma impedance changes significantly during an arc event [ 13 , 20 ]. As the reflection coefficient approaches 1, the reflected power increases nonlinearly, indicating a greater impedance mismatch or instability in the plasma, and the ideal line between the reflection coefficient and the reflected power can be calculated according to Eq. 1 . In arc management, arc generation can be addressed using a variety of methods, similar to arc suppression strategies. A common approach is to shut down the generator for a preset time when an arc is detected [ 21 ]. In this study, we use the method of shutting down the generator output for a fixed period of time when an arc is first detected. Figure 1 (a) shows the 13.56 MHz RF generator arc model with adjustment of suppression time and re-ignition time under arc conditions. Figure 1 (b) shows the forward and reflected power fluctuations that occur during an arc event. If the arc still exists after the RF power is restored, the generator will turn off again, doubling the off time, as shown in Fig. 1 (c). This incremental extension of the shutdown time continues until the arc is completely eliminated or a predefined attempt limit is reached, at which point the generator shuts down completely to protect the system. A suppression mechanism that interrupts output to stabilize the plasma and prevent system damage. If an arc is detected again during the reignition phase, the suppression time is doubled and continues to increase with each subsequent arc detection. If the problem persists, gradually increase the suppression time until the arcing problem is resolved. In addition, the system also integrates RF generator signal monitoring functions to further ensure the efficiency and reliability of the entire process. steady state. Moreover, by adjusting the reflection coefficient threshold, the system can flexibly adapt to various process requirements and achieve rapid arc extinction, especially in high-precision processes. This flexibility allows further optimization of arc management strategies to meet the stability and performance needs of both traditional and high-precision applications.In addition, the number of arcs is detected and recorded based on the RF power supply. Each time an arc event is detected, the system records the corresponding number of arcs, which helps evaluate and analyze the frequency of arc occurrence and its impact on process stability. This monitoring can more accurately adjust the suppression strategy to effectively manage the impact of arc on film quality, further improving the stability and reliability of the process. The arc event progresses through five distinct phases, shown in Fig. 2 (a). Phase I initiates the arc, where localized impedance at the arc site collapses, leading to a drop in output impedance and a rise in current to the arc burn level. Phase II follows, representing the steady-state condition, where the arc sustains a stable impedance and burns at a constant voltage and current. During this phase, unchecked arc energy may damage both the target and substrate. Phase III is the arc reaction, where the power supply cuts off power and applies a high reverse voltage, reducing arc current to zero and diverting stored energy away from the circuit. Phase IV involves shutdown, where no additional energy is applied, allowing thermal dissipation at the arc site and, in reactive processes, the reformation of an insulating layer. This period must be long enough to prevent immediate re-ignition upon power restoration, yet brief enough to maintain process stability. Phase V is the process recovery, where power is gradually re-applied by the power supply. If arc rates or plasma instability remain high, adjustments to the power supply may enhance recovery effectiveness [ 22 ]. $$\:Arc\:Energy={\int\:}_{t1}^{t2}P\left(t\right)dt={\int\:}_{t1}^{t2}v\left(t\right)i\left(t\right)dt$$ 2 Based on Eq. 2 , it defined herein is the time when the arc begin (t 1 ) and the time when the arc end (t 2 ). Selection of these times can be crucial to the arc energy calculation (time integral of V*I over the duration). In Fig. 2 (a), it marks the arc beginning time (t 1 ) and the ending time (t 2 ). In our experiments, we took a conservative approach by measuring voltage and current at the chamber instead of at the RF power supply. Measuring energy at the RF power supply output can be influenced by the reactive impedance in the transmission line and matching network. Furthermore, variations in arc energy measurements may also arise from factors such as the choice of test equipment. To ensure reliable arc energy data comparisons, it is critical to minimize discrepancies in the measurement methods [ 23 ]. 3. Experimental The chamber structure and configuration for the reactive magnetron sputtering experiments used to deposit AlN thin films in this study are illustrated in Fig. 2 (b) An AlN film was grown on a 4-inch P-type silicon substrate (100) in a high vacuum chamber, using a 4-inch diameter pure aluminum target in a high-purity nitrogen-rich atmosphere, with a substrate-to-target distance of 50 mm. Table 1 details the sputtering chamber's process parameters, including fixed pressure and nitrogen flow. RF 1.5 kW power supply (Delta Electronics Inc, Taiwan) was used to conduct experiments under identical parameters to compare their effects on AlN film sputtering. To achieve a comparable AlN film thickness, the sputtering time was set to 60 minutes. Current and voltage waveforms at the chamber connection were monitored during sputtering using an oscilloscope (Micsig TO3004, China), a high-voltage differential probe, and a 500-amp AC/DC current probe. By adjusting the reflection coefficient (Γ) range from 0.2 to 0.8 in Table 1 , analyze the impact of RF power reflection on arc suppression efficiency and determine the optimal reflection level to minimize arc energy. The reason why 0.1 and 0.9 were not selected is that the 0.9 threshold is difficult to trigger during the manufacturing process, while 0.1 is too sensitive and may cause experimental deviation, so only 0.2 to 0.8 are selected for adjustment. The RF power supply connects to the chamber via a matching box and coupler for efficient power transfer. Reactive gases (N₂, Ar) are controlled by flow meters, with a vacuum pump maintaining low pressure. The Al target acts as the cathode for sputtering, depositing AlN thin film on the substrate above. Arcing events on the target surface are monitored via an oscilloscope, while a Smith chart on a computer aids impedance matching and arc mitigation. This setup emphasizes the interplay of RF power, gas flow, and plasma dynamics. In addition, to study the effect of arc suppression time on AlN films, the arc suppression time was gradually adjusted from 2µs to 1000µs in Table 1 . The reason why we used suppression/re-ignition times of 2µs/2µs, 100µs/50µs, and 1000µs/500µs is that, according to the literature we reviewed and the practices of major manufacturers, most employ suppression/re-ignition times of 100µs/50µs and 1000µs/500µs [ 19 , 22 ]. Moreover, our RF generator is capable of setting the minimum suppression/re-ignition time to 2 µs, which enabled us to choose the 2 µs/2 µs configuration as a shorter regime to compare against the traditional, longer times. It is hypothesized that shorter arc suppression times can enhance process stability by quickly terminating arc events, while longer times can more fully dissipate the energy stored in the plasma system. At the same time, the arc re-ignition time was also adjusted in the range of 2 to 1000 µs to explore its role in the recovery of the sputtering process after an arc event. Re-ignition time is critical to balance fast recovery with avoiding immediate re-ignition. Too fast a re-ignition time can lead to repetitive arcing events and affect process stability [ 24 ]. In the RF sputtering process of AlN, the natural generation of arcs is triggered by the high current density of the plasma and the presence of reactive gases. During this process, when the current density becomes too high, the glow discharge transitions into local arc discharges, a phenomenon that cannot be completely avoided. Specifically, when RF power is too high or the gas ratio is inappropriate, the current density of the plasma concentrates on the cathode surface, causing the cathode voltage to collapse and forming local hotspots on the cathode. These hotspots initiate arc discharges, which, under high current density and heat, extinguish and reignite quickly, creating an unstable arc discharge phenomenon. Additionally, during the RF sputtering of AlN, reactive gases such as nitrogen form a dielectric layer on the cathode surface. This dielectric layer helps facilitate the transition from glow discharge to arc discharge. Due to the presence of reactive gases, the formation of this dielectric layer exacerbates arc generation, and over time, these localized arc discharges may interfere with the film deposition process and affect the quality of the AlN film. Therefore, in the RF sputtering process, the naturally occurring arcs are triggered by the interaction of high current density, localized hotspots, and reactive gases [ 25 ]. In order to ensure that the baseline conditions for adjusting parameter effects are consistent, other key conditions of the sputtering process remain fixed. Table 1 shows the process parameters for depositing aluminum nitride films using RF magnetron reactive sputtering and RF generator arc model condition, The sputtering power was set to 500 W to provide a stable and repeatable energy input, ensuring a stable deposition rate and achieving uniform film growth. The deposition time was fixed at 60 minutes to ensure sufficient film formation time while maintaining process stability. The chamber pressure was set to 5 mTorr to ensure an optimal balance between plasma density and sputtering efficiency and to avoid excessive scattering effects on film quality. Table 1 The process parameters for depositing aluminum nitride films using RF magnetron reactive sputtering and RF generator arc model condition. RFG Arc model Condition Γ(Reflection coefficient ) 0.2/0.3/0.4/0.5/0.6/0.7/0.8 Arc suppress time (S time) (µs) 2/100/1000 Arc re-ignition time (H time) (µs) 2/50/500 Sputtering Process Condition Sputtering Power(W) 500 Deposition time(min) 60 Pressure(m Torr) 5 Target to Substrate Distance (cm) 5 N 2 gas flow ratio (Sccm) 60 Ar gas flow ratio (Sccm) 15 Substrate temperature(°C) Room Temperature The target-to-substrate distance was fixed at 5 cm to standardize the plasma-substrate interaction and ensure consistent energy transfer during sputtering. In addition, the gas flow ratios of N₂ and Ar were controlled to 60 sccm and 15 sccm respectively, creating a stable nitriding environment conducive to the formation of AlN films. Maintaining this flow ratio ensures stable reaction conditions during the deposition process and promotes optimization of film structure and quality. Finally, the substrate temperature was maintained at room temperature in all experiments to more accurately observe and analyze the effects of adjusted arc-related parameters. While high temperatures typically help improve film crystallinity and adhesion, the choice of room temperature for deposition in this experiment provides a controlled baseline that helps simplify the evaluation of arc behavior and suppression efficiency. Systematic adjustments of the reflection coefficient, arc suppression time, and arc re-ignition time, while stabilizing other sputtering conditions, are essential. The determinations of the microstructure of AlN thin film were examined by X-ray diffraction (XRD), and atomic force microscope (AFM). Film thickness measurements were conducted by examining the cross-sections of the AlN films using cold field emission scanning electron microscopy (CFE-SEM, JSM-7000F, Japan) at ×50,000 magnification. Additionally, surface morphology images of the film were captured to observe the grain size at 100,000 magnifications. The morphology of the film surfaces was characterized using an atomic force microscope (AFM, Nanoview 1000, FSM, China); by using the tapping mode, the contact and lateral forces between the tip and the sample are effectively reduced. The film orientation has been determined by θ-2θ scan using an X-ray diffraction (D8 Advance, Bruker, US), and Original Pro 2018 software was used to calculate the FWHM of the 002 peak in the XRD pattern. In the measurement, A Cu target was used, and the wavelength of Cu-Kα radiation is 1.5418 Å and scanned 2θ from 30 to 40 degrees to obtain the diffraction peaks corresponding to the orientations. 4. Results and Discussion 4.1 The relationship between the reflection coefficient, arc energy and film surface property Arc counting-based detection is monitored by an RF power source, and arc energy is affected by multiple factors, including plasma stability, material properties, and chamber configuration [26]. At high arc energy, fluctuations in reflection coefficient become more pronounced, which may lead to component damage, uneven material deposition, and reduced process yield [27]. Arc energy is affected by many factors, among which transmission cables play a particularly important role in large systems [28]. Due to the inherent inductive and capacitive effects of cables, energy may gradually accumulate within the system. Theoretically, the stored energy reaches its lowest point when the cable impedance perfectly matches the load or sputter source impedance [29]. However, achieving such an ideal pairing has proven extremely challenging. This is especially true in high-power density plasma systems, where impedance mismatches often result in additional energy being released into the arc, further exacerbating system stability challenges. Importantly, an impedance mismatch may lead to an increase in stored energy, which may subsequently discharge into an arc [30]. While low arc energy is generally preferred, some extended processes can benefit from moderate arc energy levels. Typically, low-arc energy is the best choice; however, due to the rapid release of energy in the arc (see Fig. 2(a)), arc detection and response systems must act quickly to effectively limit the total energy release. Research has shown that significant amounts of energy can be released into the arc before the arc is detected by the power supply, meaning that the ability to control energy through arc response alone is limited by other system factors. These include material properties, transmission cables, generators, and filtering components in the power transmission path, all of which individually affect arc energy. Therefore, arc energy reduction through generator control alone is limited by these system-specific factors [31]. Figure 3 shows the arc energy, root mean square (RMS) surface roughness, and full width at half maximum (FWHM) on the deposition process and deposited AlN film. A systematic experimental analysis was conducted under the conditions of a fixed suppression time of 2 µs and consistency of parameters in the strategy, thereby eliminating the effects of suppression/ re-ignition time variations and focusing on the impact of the reflection coefficient on the film. When the set reflection coefficient threshold is triggered, the number of triggers is recorded. When the reflection coefficient (Г) gradually increases from 0.2 to 0.8, the arc energy shows an obvious upward trend, especially when Г = 0.7 ~ 0.8, reaching a peak value (about 0.9 mJ/kW), which shows that the increase in arc energy reflection coefficient makes the arc frequency and energy release rise simultaneously, further promoting the accumulation of arc energy. It can be seen from the RMS of the film in Fig. 3 (a) that as the reflection coefficient increases, the arc energy increases, the FWHM in Fig. 3 (b) also changes from 0.31 to 0.36, and the local impedance changes rapidly, causing the reflection coefficient to increase. Both show that at high Γ values, this results in an increase in arc energy. By adjusting the threshold value of the reflection coefficient (Г), it was observed that the arc energy has a certain impact on the quality difference of the AlN film. The trigger value settings have appropriate suppression mechanisms in different sputtering processes, thereby reducing energy accumulation. Although the triggering value is relatively sensitive under the condition of Г = 0.2 and the triggering frequency of arc events is higher, it can relatively quickly suppress small energy arcs and may accumulate to produce hard arcing. It is conducive to the stable progress of the deposition process. However, as the reflection coefficient threshold gradually increases, the reflected energy increases, causing the arc energy to gradually increase, making the system unstable, eventually triggering an arc event, and causing the energy released every time to lead the arc triggered to rise simultaneously. Therefore, when the reflection coefficient increases, the energy to trigger the arc also increases, including disturbance of film surface morphology, increased roughness, and the generation of lattice defects. It can be seen that the setting of the reflection coefficient threshold has a critical impact on the triggering of arc energy. An excessively high reflection coefficient will lead to continuous accumulation of energy and ultimately cause deterioration of film quality. Therefore, reasonable adjustment of the reflection coefficient threshold to ensure a balanced energy release can effectively reduce the adverse effects of arc activity, maintain the stability of the deposition process, and improve film quality [32]. 4.2 The effects of Γ on thickness, arc energy, and arc count Although the overall arc energy is relatively low at Γ = 0.2 due to frequent suppression, the constant triggering of the suppression mechanism significantly affects the deposition rate. This is evidenced by the thinner film thickness observed at this reflection coefficient, as shown in Fig. 4(a) and 4 (b). In Fig. 4 (b), especially for lower Γ values, the arc count increases sharply, but the arc energy is relatively low, which means that the micron-scale arc generated in the process can be detected and suppressed to stabilize plasmas and processes. A higher Γ value (0.8) means that a higher reflected power is required to trigger the threshold setting. Therefore, many micron-sized arcs will not be detected during the manufacturing process but will accumulate, eventually creating a larger arc, which will trigger to Γ = 0.8 and suppress it, thereby generating greater energy, forming a hard arc. The reduced deposition rate can be attributed to the repeated interruptions in power delivery, which reduce the effective sputtering time during the deposition process. Figure 4 (a) illustrates that the AlN film thickness increases as the reflection coefficient progresses from 0.2 to 0.8. At Γ = 0.2, the film thickness is only 182 nm, which is significantly lower compared to the 203 nm observed at Γ = 0.8. This trend indicates that higher reflection coefficients result in more stable power delivery and fewer interruptions, which promote a higher deposition rate and improved film thickness. Figure 4 (b) further highlights the relationship between reflection coefficient, arc energy, and film thickness. The arc energy is minimal at Γ = 0.2 but gradually increases with higher reflection coefficients, peaking at Γ = 0.8. In the meantime, the FWHM and RMS values, which are indicators of the film's structural quality and surface roughness, also exhibit variation with the reflection coefficient. While the arc energy increases at higher reflection coefficients, the FWHM shows an improvement, suggesting better crystallinity at these settings. However, the RMS values tend to rise slightly, indicating a trade-off between surface smoothness and structural quality (see Fig. 3c and 3d). In summary, while the reflection coefficient of Γ = 0.2 minimizes arc energy, it leads to an excessive number of arc events due to the high sensitivity of the suppression mechanism. This frequent triggering disrupts the sputtering process and reduces the deposition rate, resulting in thinner films. 4.3 The relationship between reflection coefficient, arc energy and film surface property for different suppression settings Although at a reflection coefficient of Γ = 0.2, with the power detection time fixed at 3µs for very fast detection, the number of arc events is relatively high, the energy of each arc remains low and can be classified as micro arcing [33]. In fact, due to technological advancements, we are able to trigger suppression at the onset of even the smallest arcs, which is a key factor for process stability [34]. While frequent arc events may affect the deposition rate, the lower energy per arc helps minimize the interference of high-energy arcs on the process, thereby improving film stability. This setup effectively suppresses high-energy arcs, preventing surface irregularities and structural issues. Moreover, the frequent triggering of arc suppression may lead to power output interruptions, which, in certain cases, helps regulate arc energy, benefiting the overall deposition process. Although the film thickness is thinner at Γ = 0.2, as shown in Fig. 4, the standard deviation falls within the 180–210 nm thickness range, and due to the lower arc energy, the film quality and stability are better. This indicates that it can enhance process stability and improve film quality. Therefore, a well-adjusted reflection coefficient can balance stability and deposition rate, which is crucial for producing high-quality films. The suppression time setting is critical for managing arcs and stabilizing film deposition [35]. However, overly long suppression times can increase arc energy during re-ignition, harming film quality. This study compares three suppression/re-ignition settings—2µs/2µs, 100µs/50µs, and 1ms/500µs—to highlight the advantages of ultrafast suppression [24]. As shown in Fig. 5(a), arc energy rises sharply with the reflection coefficient (Γ), especially when Γ > 0.5. With short suppression (2µs/2µs), the reflection coefficient stays between 0.2–0.5, maintaining low arc energy and better stability. In contrast, longer suppression times show greater energy fluctuations under high Γ, indicating poorer control and potential film damage. Although longer suppression reduces arc frequency, it increases the energy per event, harming plasma stability and film quality. Figure 5(b) shows that 2µs/2µs leads to more frequent but lower-energy arcs, minimizing film damage, whereas 100µs/50µs results in fewer but more damaging arcs. Thus, short suppression better protects film quality and process stability [29]. Figure 6 shows that shorter suppression (2µs/2µs) yields better FWHM, RMS, and film thickness results, confirming its positive impact on deposition quality [24, 27]. In contrast, traditional longer suppression (e.g., 100µs/50µs) often destabilizes the process and worsens film quality [27], causing increased defects, roughness, and thickness variation.Short suppression quickly extinguishes arcs, preventing excessive heating and maintaining substrate stability, thereby improving crystallinity and film quality. Longer suppression may fail to fully extinguish arcs, leading to thermal damage and film instability. Long processing times also destabilize deposition and reduce film quality. Because longer suppression increases power loss, especially at high arc frequencies, short suppression is vital for stabilizing deposition and ensuring efficient power transmission [8]. Figure 6(a) shows that at a reflection coefficient of 0.2, arc energy under 2µs/2µs is lower than under 1ms/500µs. Longer suppression leads to higher arc energy growth with increasing reflection coefficient, highlighting the need to carefully control Γ during RF plasma operation. Figure 5(b) shows that under 2µs/2µs at Γ = 0.2, there are more arcs but lower energy, effectively managing micron-sized arcs. Under 1ms/500µs, arc counts drop and energy rises with higher Γ, indicating that strong reflection inhibits arc occurrence but increases the energy of individual arcs. Figure 6 (b) shows that the short-term condition is the most sensitive to reflection, with arc counts decreasing sharply as the reflection coefficient increases, while the long-term condition shows a slower decreasing trend. The dual influence of the reflection coefficient on arc behavior is illustrated. A high reflection coefficient threshold results in high arc energy that cannot be detected and suppressed but also reduces the frequency of arcing. These results provide important parametric implications for future basic plasma deposition. Especially in the film deposition process, proper control of the reflection coefficient range and suppression time can effectively improve system stability and film quality. At the machine's fastest suppression time of 2µs/2µs, the FWHM is significantly lower, indicating that the film crystal quality is better; the RMS value is also relatively small, indicating that the film surface is smoother. In addition, the film thickness distribution is uniform across all values of Γ, indicating that the deposition rate is less affected by the arc [25]. These results illustrate the importance of balancing suppression time and reflection coefficient thresholds to optimize the response of the arc management system and the resulting film properties. Inadequate arc suppression can lead to the continuous occurrence of arcs. These persistent arcs, often referred to as "hard arcs," are essentially similar to "micro arcs," except that they may regenerate or persist beyond the initial suppression attempts. This can happen when arcs form in areas more likely to reignite or due to insufficient arc response. When arcs regenerate, it can result in a higher total count and energy released into the affected spot, increasing the risk of particles and damage [24]. By improving these ultrafast arc management strategies, stable plasma conditions and high-quality AlN films can be achieved, which are critical for the advancement of semiconductor and optoelectronic devices [35]. The results show that the combination of fast suppression times and carefully calibrated reflection coefficient thresholds can enhance the deposition process and ensure consistent film performance. 4.4 PCA Analysis of Spectra under Different Arc Suppression Times According to the experimental results shown in Fig. 7 (a) and (b), a comparison of the optical emission spectroscopy (OES) data collected under different suppression times reveals that due to the vast amount of OES data, observations were first focused on the main wavelengths associated with aluminum nitride. These wavelengths include nitrogen (336.46 nm, 315.2 nm, 379.86 nm), nitrogen ions (390.93 nm, 427.29 nm), and aluminum (356.9 nm, 374.68 nm). For a one-hour processing period, each dataset contains approximately 200,000 spectral data points. Through full-spectrum analysis, it was found that although different suppression times have some impact on system performance, the variations in spectral intensity and shape across the full spectra shown in Fig. 7 (b) are not prominent, making it difficult to directly observe any significant differences. This is particularly true when the suppression time ranges from as short as 2µs to as long as 1ms, while the maximum resolution of the OES equipment is one data acquisition every 20 milliseconds. This means the temporal resolution of the data is limited and cannot capture subtle variations at the microsecond level, resulting in very similar spectral intensities and shapes across different suppression times. Therefore, in order to more accurately extract the hidden information within the data, we introduced principal component analysis (PCA) for dimensionality reduction of the OES data. PCA can effectively extract the main features from spectral data, reducing its dimensionality while preserving key information. After applying PCA, we were able to clearly observe classification characteristics in the data corresponding to different suppression times—features that may not be easily noticeable through full-spectrum intensity or shape alone. Through PCA processing, not only can seemingly similar data be clustered, but the influence of different suppression times on the OES spectra can also be more deeply analyzed and interpreted. In addition, experiments were conducted under identical chamber conditions to ensure all variables remained consistent, thereby eliminating potential interfering factors that could affect the data. This analytical approach not only enhances the accuracy of the experimental results but also provides a more scientific basis for optimizing suppression time parameters, further improving process stability and efficiency. The application of PCA enables us to overcome the limitations of traditional full-spectrum analysis, uncovering deeper correlations between suppression time and OES data. This offers valuable insights for subsequent process optimization and technological improvements. According to Fig. 8, PCA was used to process the high-dimensional data of OES, simplifying the data structure and extracting key information. PCA identifies the principal components that account for the greatest variance in the data and projects the original high-dimensional data into a lower-dimensional space. This reduces the data's dimensionality while retaining the most important features for analysis. This method not only effectively improves data processing efficiency but also helps identify the main parameters that affect the quality of AIN films, further optimizing process conditions to ensure the stability and high-quality performance of the films. PCA identifies the directions of maximum variance in the data, called principal components, which are linear combinations of the original variables. These principal components are ranked according to the variance they explain, with the first principal component capturing the greatest variance, the second orthogonal principal component capturing the second largest variance, and so on. This enables PCA to reduce the complexity of the data while preserving its core patterns [36, 37]. These results show that balancing suppression time and reflection coefficient (Γ) thresholds is key to optimizing arc management and film properties. Insufficient suppression can cause persistent "hard arcs," which, like micro-arcs, may regenerate and increase total arc counts and energy release, raising damage risks [38, 39]. Improving ultrafast arc management strategies enables stable plasma and high-quality AlN films, essential for semiconductor and optoelectronic device development [40, 41]. The study finds that combining rapid suppression times with well-calibrated Γ thresholds enhances deposition consistency. The Γ threshold controls when arc suppression triggers, affecting process stability and film quality. A higher Γ allows more arc energy to accumulate before suppression, reducing shutdown frequency and improving operational stability, but may cause localized surface damage if energy becomes too concentrated. Conversely, a lower Γ triggers earlier suppression, reducing arc energy and events but may cause frequent shutdowns and re-ignitions, impacting stability and film quality. Thus, the optimal Γ setting must balance arc energy, suppression system stability, and film quality. A moderately high Γ can stabilize deposition while avoiding surface damage, whereas too low a Γ decreases sputtering efficiency and film uniformity by causing excessive system interruptions. A carefully chosen Γ helps maintain stable deposition while effectively managing arc energy. 5. Conclusion In the present study, we systematically demonstrate a novel methodology for rapid and precise arc detection and suppression in RF plasma sputtering systems for aluminum nitride (AlN) film deposition. By meticulously optimizing process parameters—specifically, setting the reflection coefficient at 0.2 and employing ultrafast suppression and times of 2µs we effectively mitigate high-frequency micro-arcing and maintain process stability under conditions of elevated arc occurrence. Our experimental results indicate that this configuration circumvents the limitations associated with longer suppression intervals, which can lead to energy accumulation and subsequent process instability. Consequently, this work provides critical insights into the interplay between arc suppression parameters and film quality. The key innovation of this study lies in the introduction of an ultrafast arc management strategy, which, to the best of our knowledge, is the first to systematically investigate a 2 µs/2 µs suppression/re-ignition regime in RF sputtering processes. This ultrashort modulation enables arc suppression at the micro-discharge stage, preventing the emergence of high-energy “hard arcs” and ensuring a more stable plasma environment. As a result, the approach significantly improves AlN film crystallinity, reduces surface roughness, and maintains uniform deposition rates. Additionally, by integrating optical emission spectroscopy (OES) with principal component analysis (PCA), the study contributes a data-driven framework that correlates plasma dynamics with thin film characteristics across different arc management regimes. The implications of our findings are substantial for the fabrication of microelectronic and optoelectronic devices, where maintaining consistent process conditions and achieving high-quality film deposition are essential. Moreover, this study lays the groundwork for future research endeavors aimed at further refining arc suppression strategies. Prospective investigations may include the development of adaptive real-time control schemes and a more comprehensive analysis of the relationship between arc energies and the resultant film microstructure. Such research will not only deepen the fundamental understanding of OES plasma spectra from sputtering phenomena but will also drive technological enhancements in thin film deposition processes. Declarations Acknowledgements This research is supported by Delta Electronics Taiwan (Project Number: 11313081). We sincerely appreciate the generous support from Delta Electronics, which has enabled the successful completion of this study. Funding This study was financially supported by Delta Electronics, Inc., Taiwan, and Department of Mechanical Engineering, National Central University, Taiwan (No. 11213080) Conflict of Interest The authors have no conflicts to disclose Author Contributions Yu-Shin Chen : Data curation (equal); Formal analysis (equal); Investigation (equal); Methodology (equal); Writing – original draft (equal). Cheng-Yuan Kao : Data curation (equal); Investigation (equal); Methodology (equal); Writing–original draft (equal). Hsuan-Fan Chen : Data curation (equal); Formal analysis (equal); Methodology (equal); Validation (equal); Writing–original draft (equal). Chih-Hao Tsui : Resources (equal). Ting-Yueh Yang : Resources (equal). Yiin-Kuen Fuh : Investigation (equal); Resources (equal); Supervision (equal); Writing – review & editing (equal). Tomi T. Li : Supervision (equal); Writing – original draft (equal); Writing – review & editing (equal). References Yamashita, H., Fukui, K., Misawa, S., & Yoshida, S. (1979). Optical properties of AlN epitaxial thin films in the vacuum ultraviolet region. Journal of Applied Physics, 50(2), 896-898. Slack, G. A., Tanzilli, R. A., Pohl, R. O., & Vandersande, J. W. (1987). The intrinsic thermal conductivity of AlN. Journal of Physics and Chemistry of Solids, 48(7), 641-647. Haider, S. T., Shah, M. A., Lee, D. G., & Hur, S. (2023). A review of the recent applications of aluminum nitride-based piezoelectric devices. IEEE Access, 11, 58779-58795. Ohta, J., Fujioka, H., Ito, S., & Oshima, M. (2002). Room-temperature epitaxial growth of AlN films. Applied Physics Letters, 81(13), 2373-2375. Liu, B., Gao, J., Wu, K. M., & Liu, C. (2009). Preparation and rapid thermal annealing of AlN thin films grown by molecular beam epitaxy. Solid State Communications, 149(17-18), 715-717. Khan, S., et al. (2015). Influence of the nitrogen fraction on AlN thin film deposited by cathodic arc ion. Materials Science in Semiconductor Processing, 29, 193-200. Baba, S., Gajewski, W., Jasinski, M., Zelechowski, M., & Kazmierkowski, M. P. (2021). High performance power supplies for plasma materials processing. IEEE Access, 9, 19327-19344. Carter, D. (2008). Arc prevention in magnetron sputtering processes. 51st Annual Technical Conference Proceedings of the Society of Vacuum Coaters, 380. Anders, A. (2008). Cathodic arcs: From fractal spots to energetic condensation (Vol. 50). Springer. Anders, A. (2006). Physics of arcing, and implications to sputter deposition. Thin Solid Films, 502(1-2), 22-28. Bogatin, E. (2004). Signal integrity: Simplified. Prentice Hall Professional. Belkind, A., Freilich, A., & Scholl, R. (1998). Pulse duration effects in pulse-power reactive sputtering of Al2O3. Surface and Coatings Technology, 108, 558-563. Scholl, R. A. (1997). Power systems for reactive sputtering of insulating films. Surface and Coatings Technology, 93(1), 7-13. Maniv, S., & Westwood, W. D. (1980). Discharge characteristics for magnetron sputtering of Al in Ar and Ar/O2 mixtures. Journal of Vacuum Science and Technology, 17(3), 743-751. Carter, D., Walde, H., & Nauman, K. (2012). Managing arcs in large area sputtering applications. Thin Solid Films, 520(12), 4199-4202. Samukawa, S. (1996). Pulse‐time‐modulated electron cyclotron resonance plasma etching with low radio‐frequency substrate bias. Applied Physics Letters, 68(3), 316-318. Caughman, J. B. O., & Hoffman, D. J. (1994). ARC detection and protection in high‐power antenna systems. AIP Conference Proceedings, 289(1), 379-382. Ambrosy, U. (2012). TruPlasma RF series 3000, 7000 new. Proceedings of the 3rd International Conference on Power Electronics Plasma Engineering, 14.1-14.5. Wiedemuth, P., Ambrosy, U., Mann, E., Grede, A., & Thome, C. (2013). A new RF power supply platform for plasma applications. Proceedings of the 56th Annual Technical Conference of the Society of Vacuum Coaters, 571. Pankratz, J., & Hoffman, J. (2022). Understanding arc energy: Methods to measure, compare, and control arc energy. SVC Bulletin, Summer, 2-7. Panjan, P., Drnovšek, A., Gselman, P., Čekada, M., & Panjan, M. (2020). Review of growth defects in thin films prepared by PVD techniques. Coatings, 10(5), 447. Van Zyl, G., & Heckman, R. (2005). Managing arcs in RF powered plasma processes. Proceedings of the 48th Annual Technical Conference of the Society of Vacuum Coaters, 44-49. Cho, C., Kim, S., Choi, M., Lee, Y., Seong, I., Jeong, W., ... & You, S. J. (2024). Initiation mechanism of arcing generated in RF capacitively coupled plasma. Physica Scripta, 99(5), 055603. Heintze, M., & Luciu, I. (2018). Nodule formation on sputtering targets: Causes and their control by MF power supplies. Surface and Coatings Technology, 336, 80-83. Oskirko, V. O., Semenov, V. D., Solovyev, A. A., Rabotkin, S. V., Pavlov, A. P., & Zakharov, A. N. (2022). Arc energy minimization in high-power impulse magnetron sputtering. Vacuum, 202, 111213. Guljakow, J., & Lang, W. (2024). Analysis of arcing in processes during the reactive deposition of aluminum in a High Power Impulse Magnetron Sputtering process. hal-04669476f. Tiron, V., Velicu, I. L., Matei, T., Cristea, D., Cunha, L., & Stoian, G. (2020). Ultra-short pulse HiPIMS: A strategy to suppress arcing during reactive deposition of SiO2 thin films with enhanced mechanical and optical properties. Coatings, 10(7), 633. Wiedemuth, P., Merte, R., Richter, U., & Bannwarth, M. (2012). Next generation of mid-frequency power supplies for plasma applications. Proceedings of the 55th SVC Annual Technical Conference, 195-198. Walde, H. V., Christie, D. J., Kovacevic, S., & Tomasel, F. G. (2012). Advances in power supplies for high arc rate applications. Proceedings of the 55th SVC Annual Technical Conference, 191-193. Yamada, A., Arai, H., Obayashi, S., & Shoki, H. (2011). Evaluation for RFID tag antennas by antenna clearance based on power reflection coefficient. Proceedings of the 5th European Conference on Antennas and Propagation (EUCAP), 2997-3000. Jun, H. S. (2013). Statistical prediction of micro-arcing in radio-frequency plasmas for the manufacture of semiconductor devices. Japanese Journal of Applied Physics, 52(6R), 066202. Rogers, J. H., LaRue, P., Phelps, D. A., & Pinsker, R. I. (1995). RF arc detection using harmonic signals. Proceedings of the 16th International Symposium on Fusion Engineering, 1, 522-525. Yan, E., et al. (2009). Solving arcing issues in CVD processes. ECS Transactions, 18(1), 581. Heintze, M., & Kroyer, T. (2016). Benefit of modern arc management in MF dual magnetron sputtering power supplies: Stable deposition process for a wide range of materials. Vakuum in Forschung und Praxis, 28(4), 26-29. Yin, Y., et al. (2007). Time dependent plasma properties during microarcing in radio frequency plasmas. Applied Physics Letters, 91(19), 191504. Bu, G., Ciplys, D., Shur, M., Schowalter, L. J., Schujman, S., & Gaska, R. (2006). Surface acoustic wave velocity in single-crystal AIN substrates. IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control, 53(1), 251-254. Seller, J. (1998). Asymmetric bipolar pulsed DC: The enabling technology for reactive PVD. Surface and Coatings Technology, 98(1-3), 1245-1250. Güttler, D., Grötzschel, R., & Möller, W. (2007). Lateral variation of target poisoning during reactive magnetron sputtering. Applied Physics Letters, 90(26). Ruiz, E., Alvarez, S., & Alemany, P. (1994). Electronic structure and properties of AIN. Physical Review B, 49(11), 7115. Belkind, A., Freilich, A., Lopez, J., Zhao, Z., Zhu, W., & Becker, K. (2005). Characterization of pulsed DC magnetron sputtering plasmas. New Journal of Physics, 7(1), 90. Carter, D., Walde, H., McDonough, G., & Roche, G. (2002). Parameter optimization in pulsed DC reactive sputter deposition of aluminum oxide. Proceedings of the Annual Technical Conference-Society of Vacuum Coaters, 570-577. Soltanpour, P. N., Jones, J. B., Jr., & Workman, S. M. (1982). Optical emission spectrometry. In A. L. Page (Ed.), Methods of Soil Analysis: Part 2: Chemical and Microbiological Properties (pp. 29–65). ASA. Qayyum, A., Zeb, S., Naveed, M., Ghauri, S., Waheed, A., & Zakaullah, M. (2006). Plasma Devices Operations, 14, 61. Kau, L. H., Huang, H. J., Chang, H. E., Hsieh, Y. L., Fuh, Y. K., & Li, T. T. (2020). Micro Nano Letters, 15, 323. Zhou, W. Y., Zhang, W. Z., Wang, L. S., & Zhang, H. J. (2023). Materials, 16, 3015. Hong, S. J., May, G. S., & Park, D. C. (2003). IEEE Transactions on Semiconductor Manufacturing, 16, 598. Chen, W. L., Lin, C. Y., Huang, J. M., & Chen, T. C. (2023). International Journal of Advanced Manufacturing Technology, 127, 2955 Cite Share Download PDF Status: Published Journal Publication published 24 Jul, 2025 Read the published version in The International Journal of Advanced Manufacturing Technology → Version 1 posted Editorial decision: Minor Revisions Needed 30 Jun, 2025 Reviewers agreed at journal 27 May, 2025 Reviewers invited by journal 27 May, 2025 Editor assigned by journal 27 May, 2025 First submitted to journal 25 May, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-6737015","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":462669320,"identity":"14ecff1b-bd6f-4694-9ddf-c0a5831e5cc3","order_by":0,"name":"Yu-Shin Chen","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Yu-Shin","middleName":"","lastName":"Chen","suffix":""},{"id":462669321,"identity":"5a4c4bb6-80c5-407d-a1db-cc85b9b62549","order_by":1,"name":"Cheng-Yuan Kao","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Cheng-Yuan","middleName":"","lastName":"Kao","suffix":""},{"id":462669322,"identity":"76583024-79fa-486b-b0a0-7dbc3b5779ed","order_by":2,"name":"Hsuan-Fan Chen","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Hsuan-Fan","middleName":"","lastName":"Chen","suffix":""},{"id":462669323,"identity":"985ec5dc-4135-4325-870e-82fa3a5f323e","order_by":3,"name":"Chih-Hao Tsui","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Chih-Hao","middleName":"","lastName":"Tsui","suffix":""},{"id":462669324,"identity":"1f9f3f5e-c2f9-4f8d-97e5-9ed0cf14fb05","order_by":4,"name":"Ting-Yueh Yang","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Ting-Yueh","middleName":"","lastName":"Yang","suffix":""},{"id":462669325,"identity":"179f6506-d45e-477c-b4d0-c041cd2395f8","order_by":5,"name":"Yiin-Kuen Fuh","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA2ElEQVRIiWNgGAWjYDACZgY2IGnBzyDBfADIkJAhVouEZIMEWwKIwUOMPTAtPAYgHmEtuu3Mzx78qJCQMLjd8/nVjRoLHgb2w0c34NNidpjN3LDnDFDLnbPbrHOOAR3Gk5Z2A78WHjYJ3jaJOoMbuduMc9iAWiR4zAhqkfz7D2jLjZxnxjn/iNQizdsA1sL8OLeNKC1sZtIyxyQkJG+kmTHn9knwsBH0y/nDzyTf1NhI8N1Ifvw551udHD/74WN4tSADNgkwSaxyEGD+QIrqUTAKRsEoGDkAAIXlQYMddTLYAAAAAElFTkSuQmCC","orcid":"https://orcid.org/0000-0001-5322-0718","institution":"National Central University","correspondingAuthor":true,"prefix":"","firstName":"Yiin-Kuen","middleName":"","lastName":"Fuh","suffix":""},{"id":462669326,"identity":"b9079095-fe3f-497e-8b6c-cc6a2dbd39e1","order_by":6,"name":"Tomi T. Li","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Tomi","middleName":"T.","lastName":"Li","suffix":""}],"badges":[],"createdAt":"2025-05-24 05:58:53","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6737015/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6737015/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1007/s00170-025-16119-0","type":"published","date":"2025-07-24T15:57:40+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":83666915,"identity":"09f1315d-ae94-4116-9599-060e58cdfc84","added_by":"auto","created_at":"2025-05-30 12:01:13","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":73746,"visible":true,"origin":"","legend":"\u003cp\u003e(a) illustrates the 13.56 MHz RF generator arc model with adjustments to suppress time and pulse time under arc conditions, (b) shows the forward and reflected power fluctuations occurring during an arc event, and (c) when a second arc occurs, the closing time processing system doubles the closing time of the next arc after triggering the first arc.\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-6737015/v1/cc24aa370f589ba7d33fbb83.png"},{"id":83667439,"identity":"3f6b54c4-9ba2-4786-b435-2a1ef7a1fdfd","added_by":"auto","created_at":"2025-05-30 12:09:13","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":121005,"visible":true,"origin":"","legend":"\u003cp\u003e(a) The five phases of an arc event. (b) Illustrates a schematic of a reactive magnetron sputtering chamber with a 13.56 MHz RF generator and an oscilloscope for monitoring.\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-6737015/v1/31f6bb83f49366c56c6a29f3.png"},{"id":83666923,"identity":"88980aba-1f8f-4fec-82b7-a91e962f07df","added_by":"auto","created_at":"2025-05-30 12:01:14","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":269920,"visible":true,"origin":"","legend":"\u003cp\u003e(a) The AlN (002) peak in XRD diffraction patterns from Γ 0.2 to Γ 0.8. (b) RMS roughness values ranging from 1.12 to 1.74. (c) The relationship between the reflection coefficient, RMS, and arc energy. (d) The relationship between the reflection coefficient, FWHM, and arc energy.\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-6737015/v1/95485ea716865f88766ecbc8.png"},{"id":83666919,"identity":"2701c39b-b961-4fee-8027-2dd9f0e1d60c","added_by":"auto","created_at":"2025-05-30 12:01:14","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":209332,"visible":true,"origin":"","legend":"\u003cp\u003e(a) The reflection coefficients (Γ) from 0.2 to 0.8 on various thickness and (b) The effects of Γ on thickness, arc energy, and arc count.\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-6737015/v1/b62e0dc79c1ab24590623e44.png"},{"id":83666930,"identity":"293b8a0a-bee3-4625-bd12-c5812e8ce237","added_by":"auto","created_at":"2025-05-30 12:01:14","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":60122,"visible":true,"origin":"","legend":"\u003cp\u003e(a) Relationship between reflection coefficient (Γ) and arc energy (mJ/kW) for three different suppression settings: 2μs/2μs, 100μs/50μs and 1ms/500μs. (b) The results show that arc energy increases as the reflection coefficient increases, with 100μs/50μs exhibiting generally higher arc energy across all values of Γ.\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-6737015/v1/46975ed8ca089187db3977ea.png"},{"id":83667602,"identity":"531bfecb-cfa8-473b-9570-d56887c4f1c8","added_by":"auto","created_at":"2025-05-30 12:17:14","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":62492,"visible":true,"origin":"","legend":"\u003cp\u003e(a) Relationship between reflection coefficient (Γ) and arc energy (mJ/kW) for two different suppression settings: 2μs/2μs, 100μs/50μs and 1ms/500μs. (b) The results show that arc energy increases as the reflection coefficient increases, with 1ms/500μs exhibiting generally higher arc energy across all values of Γ.\u003c/p\u003e","description":"","filename":"6.png","url":"https://assets-eu.researchsquare.com/files/rs-6737015/v1/cf329696ca91fa978d2e2be2.png"},{"id":83666928,"identity":"5d2451a8-066d-4086-8aca-8563c1d590e2","added_by":"auto","created_at":"2025-05-30 12:01:14","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":78626,"visible":true,"origin":"","legend":"\u003cp\u003e(a) OES intensity analysis under different suppression times. (b) OES full spectrum of AIN deposition with RF generator.\u003c/p\u003e","description":"","filename":"7.png","url":"https://assets-eu.researchsquare.com/files/rs-6737015/v1/7c71d3796475e818f35f2be4.png"},{"id":83666922,"identity":"d0c27577-8092-4f53-bff1-965d3ff62367","added_by":"auto","created_at":"2025-05-30 12:01:14","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":83620,"visible":true,"origin":"","legend":"\u003cp\u003eDifferent suppression time PCA classification analysis.\u003c/p\u003e","description":"","filename":"8.png","url":"https://assets-eu.researchsquare.com/files/rs-6737015/v1/a88732c1dabb3136780b50de.png"},{"id":87756847,"identity":"382a8cc6-4c23-4fa1-9afa-4be1fbec6d5b","added_by":"auto","created_at":"2025-07-28 16:09:39","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1566932,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6737015/v1/fa54a25e-2246-485b-91d7-cc045734e8f5.pdf"}],"financialInterests":"","formattedTitle":"Optical Emission Spectroscopy and Ultrafast Arc Management Strategy on the Quality of AlN Thin Films Using a 13.56 MHz RF Generator","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eAlN has garnered substantial attention as a potential material for microelectronic and optoelectronic applications, including short-wavelength light emitters. Its exceptional properties, such as a wide band gap of 6.2 eV, excellent thermal conductivity, low thermal expansion coefficient, outstanding chemical and thermal stability, high dielectric strength, and fast surface acoustic wave velocity, make high-quality AlN films highly applicable in high-frequency and high-power device technologies [\u003cspan additionalcitationids=\"CR2 CR3 CR4 CR5 CR6\" citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eArcing problems are frequently encountered in semiconductor manufacturing, particularly at low pressures, high plasma power, and high plasma density. These issues are exacerbated in nanoscale devices, which are more vulnerable to damage caused by arcing. In CVD tools, common issues include wafer-level and electrostatic chucks (ESC) arcing in high density plasma (HDP) chambers, as well as wafer-level arcing in PECVD chambers. Arcing can lead to physical damage to chamber components, contamination, and RF system malfunctions. On device wafers, arcing can cause issues such as dielectric breakdown, interconnect failures (open or short circuit), contamination, and gate oxide integrity (GOI) failure. Identifying and addressing the root causes of arcing is essential to minimize reduced mean time between failures (MTBF) and low up-time, which can negatively affect overall fab efficiency [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eArcing has long been a known issue in DC-powered plasma systems, especially in dielectric sputtering, caused by charge buildup and electrical breakdown on dielectric films deposited on targets and chamber walls [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]. To address this, DC supplies now include advanced arc management systems that detect arcs and respond by briefly cutting power or reversing output polarity [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]. In critical processes, the downtime is factored into the total process time (joule mode). Periodic pulsing or polarity reversal can also reduce arc occurrence.\u003c/p\u003e \u003cp\u003eRF power emerged as an alternative for direct sputtering of insulators without arcing [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e], but recent studies show RF systems can still experience arcs, which may be highly disruptive. Causes include charge buildup on gate electrode patterns [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e], polymer coatings [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e], hardware defects, damage to anodized layers, electrical potential differences between components [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e], and high RF power [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]. This manuscript presents recent findings on RF arc detection and management.\u003c/p\u003e \u003cp\u003eIn magnetron sputtering, arcing remains a major challenge, mainly causing particle generation. Traditional arc management detects high current or low voltage anomalies and temporarily interrupts power to extinguish arcs [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]. Although improvements have reduced arc energy, current systems cannot react fast enough to fully prevent particle formation. The generated particles are often much larger than the film thickness, leading to serious defects and quality issues. While arc management reduces particle quantity, it remains a reactive strategy that cannot prevent particle ejection at the arc\u0026rsquo;s onset [\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eArcing is also a major cause of process control and thin film quality issues in magnetron sputtering. Unlike the stable glow of sputtering plasmas, arcs involve localized, intense discharges caused by collective electron emission during breakdown events [\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e]. The concentrated energy can produce molten material and eject macroparticles several microns in size, regardless of material or gas type. These macroparticles cause surface defects, disrupt power delivery, lower deposition rates, and result in non-uniform thickness and banding. Arcs often originate from localized charge buildup on particles, impurities, or flakes on the target surface under high ion flux conditions [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eIn this study, we investigate for the first time the impact of ultrafast suppression/re-ignition times on film quality, filling a gap in the existing literature. The reason why using ultrafast suppression/re-ignition time as small as 2 \u0026micro;s lies in its ability to promptly terminate the early-stage arc formation before it evolves into a fully developed micro-arc. This rapid interruption effectively reduces localized thermal accumulation, prevents excessive energy concentration, and facilitates a more uniform energy distribution across the deposition surface. As a result, the plasma remains in a more stable and controlled regime, promoting improved film uniformity, reduced surface roughness, and enhanced overall quality. While previous studies have predominantly investigated suppression/re-ignition intervals of 10 \u0026micro;s [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e] and 50 \u0026micro;s [\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e], the potential of employing an ultra-short 2 \u0026micro;s interval has remained largely unexplored in the context of micro-arc discharge control. In this work, our experimental results demonstrate that the implementation of a 2 \u0026micro;s suppression/re-ignition strategy significantly improves film quality, indicating its superior effectiveness in mitigating micro-arc phenomena. The key advantage of ultrafast suppression/re-ignition lies in its ability to quickly respond to micro-arcs, reducing arc-induced defects and minimizing the impact on deposition rate. Effective arc management requires both fast detection and immediate power cutoff to limit arc energy. When combined with different reflection thresholds, this method enables more precise control. Overall, this approach offers a practical and theoretical advancement over conventional arc control techniques for thin film deposition.\u003c/p\u003e \u003cp\u003eOptical emission spectroscopy (OES) is a non-invasive diagnostic tool for plasma analysis, capable of detecting light signals emitted by excited ions, atoms, and molecules within the plasma.[\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e, \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e] By comparing these spectral lines with reference databases, the corresponding chemical species can be identified. This technique has been widely applied in semiconductor processes such as etching, PECVD, and PVD.[\u003cspan additionalcitationids=\"CR45\" citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e] Moreover, previous studies have indicated a correlation between spectral signals and thin film quality characteristics.[\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e, \u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e] However, OES typically generates large volumes of data, making it challenging to extract meaningful information for classification purposes. Therefore, in this study, principal component analysis (PCA) was employed to reduce the dimensionality of OES data by transforming it into a few principal components, thereby capturing the most essential features and exploring potential correlations with film quality and different arc management strategies.\u003c/p\u003e"},{"header":"2. Arc detection and arc management","content":"\u003cp\u003eArc detection is based on monitoring the reflection coefficient (\u0026Gamma;), which quantifies the ratio of reflected to forward power. When an arc occurs, abrupt plasma impedance changes cause a significant spike in \u0026Gamma;. The RF generator\u0026rsquo;s suppression mechanism is triggered once \u0026Gamma; exceeds a predefined threshold, allowing real-time adaptation to varying process conditions. Arc detection in the RF generator is triggered using the reflection coefficient, which ranges from 0 to 1. When an arc occurs, it causes a sudden change in plasma impedance, leading to a significant increase in the reflection coefficient. The RF generator continuously monitors this value, and if it exceeds a predefined adjustable threshold within the 0 to 1 range, the system identifies an arc and activates the suppression mechanism. This adjustable threshold allows for fine-tuning to accommodate different process requirements, ensuring precise and efficient arc detection and management [\u003cspan class=\"CitationRef\"\u003e19\u003c/span\u003e].\u003c/p\u003e\n\u003cdiv id=\"Equ1\" class=\"Equation\"\u003e\n \u003cdiv class=\"mathdisplay\" id=\"FileID_Equ1\" name=\"EquationSource\"\u003e$$\\:\\varGamma\\:=\\sqrt{\\frac{{P}_{re}}{{P}_{for}}}=\\frac{{z}_{L}-{z}_{0}}{{z}_{L}+{z}_{0}}=\\left|\\frac{1-VSWR}{1+VSWR}\\right|$$\u003c/div\u003e\n \u003cdiv class=\"EquationNumber\"\u003e1\u003c/div\u003e\n\u003c/div\u003e\n\u003cp\u003eBased on Eq.\u0026nbsp;(\u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e) for voltage standing wave ratio (VSWR), the reflection coefficient (\u0026Gamma;), calculated as the ratio of reflected power (P\u003csub\u003ere\u003c/sub\u003e) to forward power (P\u003csub\u003efor\u003c/sub\u003e) or derived from the load (Z\u003csub\u003eL\u003c/sub\u003e) and characteristic (Z\u003csub\u003e0\u003c/sub\u003e) impedance serves as a key indicator for arc detection, with an adjustable threshold 0\u0026ndash;1 to trigger suppression mechanisms when plasma impedance changes significantly during an arc event [\u003cspan class=\"CitationRef\"\u003e13\u003c/span\u003e, \u003cspan class=\"CitationRef\"\u003e20\u003c/span\u003e]. As the reflection coefficient approaches 1, the reflected power increases nonlinearly, indicating a greater impedance mismatch or instability in the plasma, and the ideal line between the reflection coefficient and the reflected power can be calculated according to Eq. \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e.\u003c/p\u003e\n\u003cp\u003eIn arc management, arc generation can be addressed using a variety of methods, similar to arc suppression strategies. A common approach is to shut down the generator for a preset time when an arc is detected [\u003cspan class=\"CitationRef\"\u003e21\u003c/span\u003e]. In this study, we use the method of shutting down the generator output for a fixed period of time when an arc is first detected. Figure \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e (a) shows the 13.56 MHz RF generator arc model with adjustment of suppression time and re-ignition time under arc conditions. Figure \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e (b) shows the forward and reflected power fluctuations that occur during an arc event. If the arc still exists after the RF power is restored, the generator will turn off again, doubling the off time, as shown in Fig. \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e (c). This incremental extension of the shutdown time continues until the arc is completely eliminated or a predefined attempt limit is reached, at which point the generator shuts down completely to protect the system. A suppression mechanism that interrupts output to stabilize the plasma and prevent system damage. If an arc is detected again during the reignition phase, the suppression time is doubled and continues to increase with each subsequent arc detection. If the problem persists, gradually increase the suppression time until the arcing problem is resolved. In addition, the system also integrates RF generator signal monitoring functions to further ensure the efficiency and reliability of the entire process. steady state. Moreover, by adjusting the reflection coefficient threshold, the system can flexibly adapt to various process requirements and achieve rapid arc extinction, especially in high-precision processes. This flexibility allows further optimization of arc management strategies to meet the stability and performance needs of both traditional and high-precision applications.In addition, the number of arcs is detected and recorded based on the RF power supply. Each time an arc event is detected, the system records the corresponding number of arcs, which helps evaluate and analyze the frequency of arc occurrence and its impact on process stability. This monitoring can more accurately adjust the suppression strategy to effectively manage the impact of arc on film quality, further improving the stability and reliability of the process.\u003c/p\u003e\n\u003cp\u003eThe arc event progresses through five distinct phases, shown in Fig. \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003e (a). Phase I initiates the arc, where localized impedance at the arc site collapses, leading to a drop in output impedance and a rise in current to the arc burn level. Phase II follows, representing the steady-state condition, where the arc sustains a stable impedance and burns at a constant voltage and current. During this phase, unchecked arc energy may damage both the target and substrate. Phase III is the arc reaction, where the power supply cuts off power and applies a high reverse voltage, reducing arc current to zero and diverting stored energy away from the circuit. Phase IV involves shutdown, where no additional energy is applied, allowing thermal dissipation at the arc site and, in reactive processes, the reformation of an insulating layer. This period must be long enough to prevent immediate re-ignition upon power restoration, yet brief enough to maintain process stability. Phase V is the process recovery, where power is gradually re-applied by the power supply. If arc rates or plasma instability remain high, adjustments to the power supply may enhance recovery effectiveness [\u003cspan class=\"CitationRef\"\u003e22\u003c/span\u003e].\u003c/p\u003e\n\u003cdiv id=\"Equ2\" class=\"Equation\"\u003e\n \u003cdiv class=\"mathdisplay\" id=\"FileID_Equ2\" name=\"EquationSource\"\u003e$$\\:Arc\\:Energy={\\int\\:}_{t1}^{t2}P\\left(t\\right)dt={\\int\\:}_{t1}^{t2}v\\left(t\\right)i\\left(t\\right)dt$$\u003c/div\u003e\n \u003cdiv class=\"EquationNumber\"\u003e2\u003c/div\u003e\n\u003c/div\u003e\n\u003cp\u003eBased on Eq. \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003e, it defined herein is the time when the arc begin (t\u003csub\u003e1\u003c/sub\u003e) and the time when the arc end (t\u003csub\u003e2\u003c/sub\u003e). Selection of these times can be crucial to the arc energy calculation (time integral of V*I over the duration). In Fig. \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003e (a), it marks the arc beginning time (t\u003csub\u003e1\u003c/sub\u003e) and the ending time (t\u003csub\u003e2\u003c/sub\u003e). In our experiments, we took a conservative approach by measuring voltage and current at the chamber instead of at the RF power supply. Measuring energy at the RF power supply output can be influenced by the reactive impedance in the transmission line and matching network. Furthermore, variations in arc energy measurements may also arise from factors such as the choice of test equipment. To ensure reliable arc energy data comparisons, it is critical to minimize discrepancies in the measurement methods [\u003cspan class=\"CitationRef\"\u003e23\u003c/span\u003e].\u003c/p\u003e"},{"header":"3. Experimental","content":"\u003cp\u003eThe chamber structure and configuration for the reactive magnetron sputtering experiments used to deposit AlN thin films in this study are illustrated in Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e (b) An AlN film was grown on a 4-inch P-type silicon substrate (100) in a high vacuum chamber, using a 4-inch diameter pure aluminum target in a high-purity nitrogen-rich atmosphere, with a substrate-to-target distance of 50 mm. Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e details the sputtering chamber's process parameters, including fixed pressure and nitrogen flow. RF 1.5 kW power supply (Delta Electronics Inc, Taiwan) was used to conduct experiments under identical parameters to compare their effects on AlN film sputtering. To achieve a comparable AlN film thickness, the sputtering time was set to 60 minutes. Current and voltage waveforms at the chamber connection were monitored during sputtering using an oscilloscope (Micsig TO3004, China), a high-voltage differential probe, and a 500-amp AC/DC current probe.\u003c/p\u003e \u003cp\u003eBy adjusting the reflection coefficient (Γ) range from 0.2 to 0.8 in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e, analyze the impact of RF power reflection on arc suppression efficiency and determine the optimal reflection level to minimize arc energy. The reason why 0.1 and 0.9 were not selected is that the 0.9 threshold is difficult to trigger during the manufacturing process, while 0.1 is too sensitive and may cause experimental deviation, so only 0.2 to 0.8 are selected for adjustment. The RF power supply connects to the chamber via a matching box and coupler for efficient power transfer. Reactive gases (N₂, Ar) are controlled by flow meters, with a vacuum pump maintaining low pressure. The Al target acts as the cathode for sputtering, depositing AlN thin film on the substrate above. Arcing events on the target surface are monitored via an oscilloscope, while a Smith chart on a computer aids impedance matching and arc mitigation. This setup emphasizes the interplay of RF power, gas flow, and plasma dynamics.\u003c/p\u003e \u003cp\u003eIn addition, to study the effect of arc suppression time on AlN films, the arc suppression time was gradually adjusted from 2\u0026micro;s to 1000\u0026micro;s in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e. The reason why we used suppression/re-ignition times of 2\u0026micro;s/2\u0026micro;s, 100\u0026micro;s/50\u0026micro;s, and 1000\u0026micro;s/500\u0026micro;s is that, according to the literature we reviewed and the practices of major manufacturers, most employ suppression/re-ignition times of 100\u0026micro;s/50\u0026micro;s and 1000\u0026micro;s/500\u0026micro;s [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e]. Moreover, our RF generator is capable of setting the minimum suppression/re-ignition time to 2 \u0026micro;s, which enabled us to choose the 2 \u0026micro;s/2 \u0026micro;s configuration as a shorter regime to compare against the traditional, longer times. It is hypothesized that shorter arc suppression times can enhance process stability by quickly terminating arc events, while longer times can more fully dissipate the energy stored in the plasma system. At the same time, the arc re-ignition time was also adjusted in the range of 2 to 1000 \u0026micro;s to explore its role in the recovery of the sputtering process after an arc event. Re-ignition time is critical to balance fast recovery with avoiding immediate re-ignition. Too fast a re-ignition time can lead to repetitive arcing events and affect process stability [\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eIn the RF sputtering process of AlN, the natural generation of arcs is triggered by the high current density of the plasma and the presence of reactive gases. During this process, when the current density becomes too high, the glow discharge transitions into local arc discharges, a phenomenon that cannot be completely avoided. Specifically, when RF power is too high or the gas ratio is inappropriate, the current density of the plasma concentrates on the cathode surface, causing the cathode voltage to collapse and forming local hotspots on the cathode. These hotspots initiate arc discharges, which, under high current density and heat, extinguish and reignite quickly, creating an unstable arc discharge phenomenon. Additionally, during the RF sputtering of AlN, reactive gases such as nitrogen form a dielectric layer on the cathode surface. This dielectric layer helps facilitate the transition from glow discharge to arc discharge. Due to the presence of reactive gases, the formation of this dielectric layer exacerbates arc generation, and over time, these localized arc discharges may interfere with the film deposition process and affect the quality of the AlN film. Therefore, in the RF sputtering process, the naturally occurring arcs are triggered by the interaction of high current density, localized hotspots, and reactive gases [\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eIn order to ensure that the baseline conditions for adjusting parameter effects are consistent, other key conditions of the sputtering process remain fixed. Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e shows the process parameters for depositing aluminum nitride films using RF magnetron reactive sputtering and RF generator arc model condition, The sputtering power was set to 500 W to provide a stable and repeatable energy input, ensuring a stable deposition rate and achieving uniform film growth. The deposition time was fixed at 60 minutes to ensure sufficient film formation time while maintaining process stability. The chamber pressure was set to 5 mTorr to ensure an optimal balance between plasma density and sputtering efficiency and to avoid excessive scattering effects on film quality.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eThe process parameters for depositing aluminum nitride films using RF magnetron reactive sputtering and RF generator arc model condition.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"2\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRFG Arc model Condition\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eΓ(Reflection coefficient )\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.2/0.3/0.4/0.5/0.6/0.7/0.8\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eArc suppress time (S time) (\u0026micro;s)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2/100/1000\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eArc re-ignition time (H time) (\u0026micro;s) \u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2/50/500\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eSputtering Process Condition\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSputtering Power(W)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e500\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDeposition time(min)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e60\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePressure(m Torr)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e5\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTarget to Substrate Distance (cm)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e5\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eN\u003csub\u003e2\u003c/sub\u003e gas flow ratio (Sccm)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e60\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAr gas flow ratio (Sccm)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e15\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSubstrate temperature(\u0026deg;C)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eRoom Temperature\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eThe target-to-substrate distance was fixed at 5 cm to standardize the plasma-substrate interaction and ensure consistent energy transfer during sputtering. In addition, the gas flow ratios of N₂ and Ar were controlled to 60 sccm and 15 sccm respectively, creating a stable nitriding environment conducive to the formation of AlN films. Maintaining this flow ratio ensures stable reaction conditions during the deposition process and promotes optimization of film structure and quality. Finally, the substrate temperature was maintained at room temperature in all experiments to more accurately observe and analyze the effects of adjusted arc-related parameters. While high temperatures typically help improve film crystallinity and adhesion, the choice of room temperature for deposition in this experiment provides a controlled baseline that helps simplify the evaluation of arc behavior and suppression efficiency. Systematic adjustments of the reflection coefficient, arc suppression time, and arc re-ignition time, while stabilizing other sputtering conditions, are essential.\u003c/p\u003e \u003cp\u003eThe determinations of the microstructure of AlN thin film were examined by X-ray diffraction (XRD), and atomic force microscope (AFM). Film thickness measurements were conducted by examining the cross-sections of the AlN films using cold field emission scanning electron microscopy (CFE-SEM, JSM-7000F, Japan) at \u0026times;50,000 magnification. Additionally, surface morphology images of the film were captured to observe the grain size at 100,000 magnifications. The morphology of the film surfaces was characterized using an atomic force microscope (AFM, Nanoview 1000, FSM, China); by using the tapping mode, the contact and lateral forces between the tip and the sample are effectively reduced. The film orientation has been determined by θ-2θ scan using an X-ray diffraction (D8 Advance, Bruker, US), and Original Pro 2018 software was used to calculate the FWHM of the 002 peak in the XRD pattern. In the measurement, A Cu target was used, and the wavelength of Cu-Kα radiation is 1.5418 \u0026Aring; and scanned 2θ from 30 to 40 degrees to obtain the diffraction peaks corresponding to the orientations.\u003c/p\u003e"},{"header":"4. Results and Discussion","content":"\u003cdiv id=\"Sec5\"\u003e\n \u003ch2\u003e4.1 The relationship between the reflection coefficient, arc energy and film surface property\u003c/h2\u003e\n \u003cp\u003eArc counting-based detection is monitored by an RF power source, and arc energy is affected by multiple factors, including plasma stability, material properties, and chamber configuration [26]. At high arc energy, fluctuations in reflection coefficient become more pronounced, which may lead to component damage, uneven material deposition, and reduced process yield [27]. Arc energy is affected by many factors, among which transmission cables play a particularly important role in large systems [28]. Due to the inherent inductive and capacitive effects of cables, energy may gradually accumulate within the system. Theoretically, the stored energy reaches its lowest point when the cable impedance perfectly matches the load or sputter source impedance [29]. However, achieving such an ideal pairing has proven extremely challenging. This is especially true in high-power density plasma systems, where impedance mismatches often result in additional energy being released into the arc, further exacerbating system stability challenges. Importantly, an impedance mismatch may lead to an increase in stored energy, which may subsequently discharge into an arc [30]. While low arc energy is generally preferred, some extended processes can benefit from moderate arc energy levels. Typically, low-arc energy is the best choice; however, due to the rapid release of energy in the arc (see Fig. 2(a)), arc detection and response systems must act quickly to effectively limit the total energy release. Research has shown that significant amounts of energy can be released into the arc before the arc is detected by the power supply, meaning that the ability to control energy through arc response alone is limited by other system factors. These include material properties, transmission cables, generators, and filtering components in the power transmission path, all of which individually affect arc energy. Therefore, arc energy reduction through generator control alone is limited by these system-specific factors [31].\u003c/p\u003e\n \u003cp\u003eFigure 3 shows the arc energy, root mean square (RMS) surface roughness, and full width at half maximum (FWHM) on the deposition process and deposited AlN film. A systematic experimental analysis was conducted under the conditions of a fixed suppression time of 2 \u0026micro;s and consistency of parameters in the strategy, thereby eliminating the effects of suppression/ re-ignition time variations and focusing on the impact of the reflection coefficient on the film. When the set reflection coefficient threshold is triggered, the number of triggers is recorded. When the reflection coefficient (Г) gradually increases from 0.2 to 0.8, the arc energy shows an obvious upward trend, especially when Г = 0.7\u0026thinsp;~\u0026thinsp;0.8, reaching a peak value (about 0.9 mJ/kW), which shows that the increase in arc energy reflection coefficient makes the arc frequency and energy release rise simultaneously, further promoting the accumulation of arc energy. It can be seen from the RMS of the film in Fig. 3 (a) that as the reflection coefficient increases, the arc energy increases, the FWHM in Fig. 3 (b) also changes from 0.31 to 0.36, and the local impedance changes rapidly, causing the reflection coefficient to increase. Both show that at high \u0026Gamma; values, this results in an increase in arc energy. By adjusting the threshold value of the reflection coefficient (Г), it was observed that the arc energy has a certain impact on the quality difference of the AlN film. The trigger value settings have appropriate suppression mechanisms in different sputtering processes, thereby reducing energy accumulation. Although the triggering value is relatively sensitive under the condition of Г = 0.2 and the triggering frequency of arc events is higher, it can relatively quickly suppress small energy arcs and may accumulate to produce hard arcing. It is conducive to the stable progress of the deposition process. However, as the reflection coefficient threshold gradually increases, the reflected energy increases, causing the arc energy to gradually increase, making the system unstable, eventually triggering an arc event, and causing the energy released every time to lead the arc triggered to rise simultaneously. Therefore, when the reflection coefficient increases, the energy to trigger the arc also increases, including disturbance of film surface morphology, increased roughness, and the generation of lattice defects. It can be seen that the setting of the reflection coefficient threshold has a critical impact on the triggering of arc energy. An excessively high reflection coefficient will lead to continuous accumulation of energy and ultimately cause deterioration of film quality. Therefore, reasonable adjustment of the reflection coefficient threshold to ensure a balanced energy release can effectively reduce the adverse effects of arc activity, maintain the stability of the deposition process, and improve film quality [32].\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec6\"\u003e\n \u003ch2\u003e4.2 The effects of \u0026Gamma; on thickness, arc energy, and arc count\u003c/h2\u003e\n \u003cp\u003eAlthough the overall arc energy is relatively low at \u0026Gamma;\u0026thinsp;=\u0026thinsp;0.2 due to frequent suppression, the constant triggering of the suppression mechanism significantly affects the deposition rate. This is evidenced by the thinner film thickness observed at this reflection coefficient, as shown in Fig. 4(a) and 4 (b). In Fig. 4 (b), especially for lower \u0026Gamma; values, the arc count increases sharply, but the arc energy is relatively low, which means that the micron-scale arc generated in the process can be detected and suppressed to stabilize plasmas and processes. A higher \u0026Gamma; value (0.8) means that a higher reflected power is required to trigger the threshold setting. Therefore, many micron-sized arcs will not be detected during the manufacturing process but will accumulate, eventually creating a larger arc, which will trigger to \u0026Gamma;\u0026thinsp;=\u0026thinsp;0.8 and suppress it, thereby generating greater energy, forming a hard arc. The reduced deposition rate can be attributed to the repeated interruptions in power delivery, which reduce the effective sputtering time during the deposition process. Figure 4 (a) illustrates that the AlN film thickness increases as the reflection coefficient progresses from 0.2 to 0.8. At \u0026Gamma;\u0026thinsp;=\u0026thinsp;0.2, the film thickness is only 182 nm, which is significantly lower compared to the 203 nm observed at \u0026Gamma;\u0026thinsp;=\u0026thinsp;0.8. This trend indicates that higher reflection coefficients result in more stable power delivery and fewer interruptions, which promote a higher deposition rate and improved film thickness. Figure 4 (b) further highlights the relationship between reflection coefficient, arc energy, and film thickness. The arc energy is minimal at \u0026Gamma;\u0026thinsp;=\u0026thinsp;0.2 but gradually increases with higher reflection coefficients, peaking at \u0026Gamma;\u0026thinsp;=\u0026thinsp;0.8. In the meantime, the FWHM and RMS values, which are indicators of the film\u0026apos;s structural quality and surface roughness, also exhibit variation with the reflection coefficient. While the arc energy increases at higher reflection coefficients, the FWHM shows an improvement, suggesting better crystallinity at these settings. However, the RMS values tend to rise slightly, indicating a trade-off between surface smoothness and structural quality (see Fig. 3c and 3d). In summary, while the reflection coefficient of \u0026Gamma;\u0026thinsp;=\u0026thinsp;0.2 minimizes arc energy, it leads to an excessive number of arc events due to the high sensitivity of the suppression mechanism. This frequent triggering disrupts the sputtering process and reduces the deposition rate, resulting in thinner films.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec7\"\u003e\n \u003ch2\u003e4.3 The relationship between reflection coefficient, arc energy and film surface property for different suppression settings\u003c/h2\u003e\n \u003cp\u003eAlthough at a reflection coefficient of \u0026Gamma;\u0026thinsp;=\u0026thinsp;0.2, with the power detection time fixed at 3\u0026micro;s for very fast detection, the number of arc events is relatively high, the energy of each arc remains low and can be classified as micro arcing [33]. In fact, due to technological advancements, we are able to trigger suppression at the onset of even the smallest arcs, which is a key factor for process stability [34]. While frequent arc events may affect the deposition rate, the lower energy per arc helps minimize the interference of high-energy arcs on the process, thereby improving film stability. This setup effectively suppresses high-energy arcs, preventing surface irregularities and structural issues. Moreover, the frequent triggering of arc suppression may lead to power output interruptions, which, in certain cases, helps regulate arc energy, benefiting the overall deposition process. Although the film thickness is thinner at \u0026Gamma;\u0026thinsp;=\u0026thinsp;0.2, as shown in Fig. 4, the standard deviation falls within the 180\u0026ndash;210 nm thickness range, and due to the lower arc energy, the film quality and stability are better. This indicates that it can enhance process stability and improve film quality. Therefore, a well-adjusted reflection coefficient can balance stability and deposition rate, which is crucial for producing high-quality films.\u003c/p\u003e\n \u003cp\u003eThe suppression time setting is critical for managing arcs and stabilizing film deposition [35]. However, overly long suppression times can increase arc energy during re-ignition, harming film quality. This study compares three suppression/re-ignition settings\u0026mdash;2\u0026micro;s/2\u0026micro;s, 100\u0026micro;s/50\u0026micro;s, and 1ms/500\u0026micro;s\u0026mdash;to highlight the advantages of ultrafast suppression [24]. As shown in Fig.\u0026nbsp;5(a), arc energy rises sharply with the reflection coefficient (\u0026Gamma;), especially when \u0026Gamma;\u0026thinsp;\u0026gt;\u0026thinsp;0.5. With short suppression (2\u0026micro;s/2\u0026micro;s), the reflection coefficient stays between 0.2\u0026ndash;0.5, maintaining low arc energy and better stability. In contrast, longer suppression times show greater energy fluctuations under high \u0026Gamma;, indicating poorer control and potential film damage.\u003c/p\u003e\n \u003cp\u003eAlthough longer suppression reduces arc frequency, it increases the energy per event, harming plasma stability and film quality. Figure\u0026nbsp;5(b) shows that 2\u0026micro;s/2\u0026micro;s leads to more frequent but lower-energy arcs, minimizing film damage, whereas 100\u0026micro;s/50\u0026micro;s results in fewer but more damaging arcs. Thus, short suppression better protects film quality and process stability [29].\u003c/p\u003e\n \u003cp\u003eFigure 6 shows that shorter suppression (2\u0026micro;s/2\u0026micro;s) yields better FWHM, RMS, and film thickness results, confirming its positive impact on deposition quality [24, 27]. In contrast, traditional longer suppression (e.g., 100\u0026micro;s/50\u0026micro;s) often destabilizes the process and worsens film quality [27], causing increased defects, roughness, and thickness variation.Short suppression quickly extinguishes arcs, preventing excessive heating and maintaining substrate stability, thereby improving crystallinity and film quality. Longer suppression may fail to fully extinguish arcs, leading to thermal damage and film instability. Long processing times also destabilize deposition and reduce film quality. Because longer suppression increases power loss, especially at high arc frequencies, short suppression is vital for stabilizing deposition and ensuring efficient power transmission [8].\u003c/p\u003e\n \u003cp\u003eFigure 6(a) shows that at a reflection coefficient of 0.2, arc energy under 2\u0026micro;s/2\u0026micro;s is lower than under 1ms/500\u0026micro;s. Longer suppression leads to higher arc energy growth with increasing reflection coefficient, highlighting the need to carefully control \u0026Gamma; during RF plasma operation.\u003c/p\u003e\n \u003cp\u003eFigure 5(b) shows that under 2\u0026micro;s/2\u0026micro;s at \u0026Gamma;\u0026thinsp;=\u0026thinsp;0.2, there are more arcs but lower energy, effectively managing micron-sized arcs. Under 1ms/500\u0026micro;s, arc counts drop and energy rises with higher \u0026Gamma;, indicating that strong reflection inhibits arc occurrence but increases the energy of individual arcs.\u003c/p\u003e\n \u003cp\u003eFigure 6 (b) shows that the short-term condition is the most sensitive to reflection, with arc counts decreasing sharply as the reflection coefficient increases, while the long-term condition shows a slower decreasing trend. The dual influence of the reflection coefficient on arc behavior is illustrated. A high reflection coefficient threshold results in high arc energy that cannot be detected and suppressed but also reduces the frequency of arcing. These results provide important parametric implications for future basic plasma deposition. Especially in the film deposition process, proper control of the reflection coefficient range and suppression time can effectively improve system stability and film quality. At the machine\u0026apos;s fastest suppression time of 2\u0026micro;s/2\u0026micro;s, the FWHM is significantly lower, indicating that the film crystal quality is better; the RMS value is also relatively small, indicating that the film surface is smoother. In addition, the film thickness distribution is uniform across all values of \u0026Gamma;, indicating that the deposition rate is less affected by the arc [25]. These results illustrate the importance of balancing suppression time and reflection coefficient thresholds to optimize the response of the arc management system and the resulting film properties. Inadequate arc suppression can lead to the continuous occurrence of arcs. These persistent arcs, often referred to as \u0026quot;hard arcs,\u0026quot; are essentially similar to \u0026quot;micro arcs,\u0026quot; except that they may regenerate or persist beyond the initial suppression attempts. This can happen when arcs form in areas more likely to reignite or due to insufficient arc response. When arcs regenerate, it can result in a higher total count and energy released into the affected spot, increasing the risk of particles and damage [24]. By improving these ultrafast arc management strategies, stable plasma conditions and high-quality AlN films can be achieved, which are critical for the advancement of semiconductor and optoelectronic devices [35]. The results show that the combination of fast suppression times and carefully calibrated reflection coefficient thresholds can enhance the deposition process and ensure consistent film performance.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec8\"\u003e\n \u003ch2\u003e4.4 PCA Analysis of Spectra under Different Arc Suppression Times\u003c/h2\u003e\n \u003cp\u003eAccording to the experimental results shown in Fig. 7 (a) and (b), a comparison of the optical emission spectroscopy (OES) data collected under different suppression times reveals that due to the vast amount of OES data, observations were first focused on the main wavelengths associated with aluminum nitride. These wavelengths include nitrogen (336.46 nm, 315.2 nm, 379.86 nm), nitrogen ions (390.93 nm, 427.29 nm), and aluminum (356.9 nm, 374.68 nm). For a one-hour processing period, each dataset contains approximately 200,000 spectral data points.\u003c/p\u003e\n \u003cp\u003eThrough full-spectrum analysis, it was found that although different suppression times have some impact on system performance, the variations in spectral intensity and shape across the full spectra shown in Fig. 7 (b) are not prominent, making it difficult to directly observe any significant differences. This is particularly true when the suppression time ranges from as short as 2\u0026micro;s to as long as 1ms, while the maximum resolution of the OES equipment is one data acquisition every 20 milliseconds. This means the temporal resolution of the data is limited and cannot capture subtle variations at the microsecond level, resulting in very similar spectral intensities and shapes across different suppression times.\u003c/p\u003e\n \u003cp\u003eTherefore, in order to more accurately extract the hidden information within the data, we introduced principal component analysis (PCA) for dimensionality reduction of the OES data. PCA can effectively extract the main features from spectral data, reducing its dimensionality while preserving key information. After applying PCA, we were able to clearly observe classification characteristics in the data corresponding to different suppression times\u0026mdash;features that may not be easily noticeable through full-spectrum intensity or shape alone. Through PCA processing, not only can seemingly similar data be clustered, but the influence of different suppression times on the OES spectra can also be more deeply analyzed and interpreted.\u003c/p\u003e\n \u003cp\u003eIn addition, experiments were conducted under identical chamber conditions to ensure all variables remained consistent, thereby eliminating potential interfering factors that could affect the data. This analytical approach not only enhances the accuracy of the experimental results but also provides a more scientific basis for optimizing suppression time parameters, further improving process stability and efficiency. The application of PCA enables us to overcome the limitations of traditional full-spectrum analysis, uncovering deeper correlations between suppression time and OES data. This offers valuable insights for subsequent process optimization and technological improvements.\u003c/p\u003e\n \u003cp\u003eAccording to Fig. 8, PCA was used to process the high-dimensional data of OES, simplifying the data structure and extracting key information. PCA identifies the principal components that account for the greatest variance in the data and projects the original high-dimensional data into a lower-dimensional space. This reduces the data\u0026apos;s dimensionality while retaining the most important features for analysis. This method not only effectively improves data processing efficiency but also helps identify the main parameters that affect the quality of AIN films, further optimizing process conditions to ensure the stability and high-quality performance of the films. PCA identifies the directions of maximum variance in the data, called principal components, which are linear combinations of the original variables. These principal components are ranked according to the variance they explain, with the first principal component capturing the greatest variance, the second orthogonal principal component capturing the second largest variance, and so on. This enables PCA to reduce the complexity of the data while preserving its core patterns [36, 37].\u003c/p\u003e\n \u003cp\u003eThese results show that balancing suppression time and reflection coefficient (\u0026Gamma;) thresholds is key to optimizing arc management and film properties. Insufficient suppression can cause persistent \u0026quot;hard arcs,\u0026quot; which, like micro-arcs, may regenerate and increase total arc counts and energy release, raising damage risks [38, 39]. Improving ultrafast arc management strategies enables stable plasma and high-quality AlN films, essential for semiconductor and optoelectronic device development [40, 41].\u003c/p\u003e\n \u003cp\u003eThe study finds that combining rapid suppression times with well-calibrated \u0026Gamma; thresholds enhances deposition consistency. The \u0026Gamma; threshold controls when arc suppression triggers, affecting process stability and film quality. A higher \u0026Gamma; allows more arc energy to accumulate before suppression, reducing shutdown frequency and improving operational stability, but may cause localized surface damage if energy becomes too concentrated.\u003c/p\u003e\n \u003cp\u003eConversely, a lower \u0026Gamma; triggers earlier suppression, reducing arc energy and events but may cause frequent shutdowns and re-ignitions, impacting stability and film quality.\u003c/p\u003e\n \u003cp\u003eThus, the optimal \u0026Gamma; setting must balance arc energy, suppression system stability, and film quality. A moderately high \u0026Gamma; can stabilize deposition while avoiding surface damage, whereas too low a \u0026Gamma; decreases sputtering efficiency and film uniformity by causing excessive system interruptions. A carefully chosen \u0026Gamma; helps maintain stable deposition while effectively managing arc energy.\u003c/p\u003e\n\u003c/div\u003e"},{"header":"5. Conclusion","content":"\u003cp\u003eIn the present study, we systematically demonstrate a novel methodology for rapid and precise arc detection and suppression in RF plasma sputtering systems for aluminum nitride (AlN) film deposition. By meticulously optimizing process parameters\u0026mdash;specifically, setting the reflection coefficient at 0.2 and employing ultrafast suppression and times of 2\u0026micro;s we effectively mitigate high-frequency micro-arcing and maintain process stability under conditions of elevated arc occurrence. Our experimental results indicate that this configuration circumvents the limitations associated with longer suppression intervals, which can lead to energy accumulation and subsequent process instability. Consequently, this work provides critical insights into the interplay between arc suppression parameters and film quality.\u003c/p\u003e \u003cp\u003eThe key innovation of this study lies in the introduction of an ultrafast arc management strategy, which, to the best of our knowledge, is the first to systematically investigate a 2 \u0026micro;s/2 \u0026micro;s suppression/re-ignition regime in RF sputtering processes. This ultrashort modulation enables arc suppression at the micro-discharge stage, preventing the emergence of high-energy \u0026ldquo;hard arcs\u0026rdquo; and ensuring a more stable plasma environment. As a result, the approach significantly improves AlN film crystallinity, reduces surface roughness, and maintains uniform deposition rates. Additionally, by integrating optical emission spectroscopy (OES) with principal component analysis (PCA), the study contributes a data-driven framework that correlates plasma dynamics with thin film characteristics across different arc management regimes.\u003c/p\u003e \u003cp\u003eThe implications of our findings are substantial for the fabrication of microelectronic and optoelectronic devices, where maintaining consistent process conditions and achieving high-quality film deposition are essential. Moreover, this study lays the groundwork for future research endeavors aimed at further refining arc suppression strategies. Prospective investigations may include the development of adaptive real-time control schemes and a more comprehensive analysis of the relationship between arc energies and the resultant film microstructure. Such research will not only deepen the fundamental understanding of OES plasma spectra from sputtering phenomena but will also drive technological enhancements in thin film deposition processes.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis research is supported by Delta Electronics Taiwan (Project Number: 11313081). We sincerely appreciate the generous support from Delta Electronics, which has enabled the successful completion of this study.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study was financially supported by Delta Electronics, Inc., Taiwan, and Department of Mechanical Engineering, National Central University, Taiwan (No. 11213080)\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConflict of Interest\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors have no conflicts to disclose\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor Contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eYu-Shin Chen\u003c/strong\u003e: Data curation (equal); Formal analysis (equal); Investigation (equal); Methodology (equal); Writing – original draft (equal). \u003cstrong\u003eCheng-Yuan Kao\u003c/strong\u003e: Data curation (equal); Investigation (equal); Methodology (equal); Writing–original draft (equal). \u003cstrong\u003eHsuan-Fan Chen\u003c/strong\u003e: Data curation (equal); Formal analysis (equal); Methodology (equal); Validation (equal); Writing–original draft (equal). \u003cstrong\u003eChih-Hao Tsui\u003c/strong\u003e: Resources (equal). \u003cstrong\u003eTing-Yueh Yang\u003c/strong\u003e: Resources (equal). \u003cstrong\u003eYiin-Kuen Fuh\u003c/strong\u003e: Investigation (equal); Resources (equal); Supervision (equal); Writing – review \u0026amp; editing (equal).\u003cstrong\u003eTomi T. Li\u003c/strong\u003e: Supervision (equal); Writing – original draft (equal); Writing – review \u0026amp; editing (equal).\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eYamashita, H., Fukui, K., Misawa, S., \u0026amp; Yoshida, S. (1979). Optical properties of AlN epitaxial thin films in the vacuum ultraviolet region. Journal of Applied Physics, 50(2), 896-898.\u003c/li\u003e\n\u003cli\u003eSlack, G. A., Tanzilli, R. A., Pohl, R. O., \u0026amp; Vandersande, J. W. (1987). The intrinsic thermal conductivity of AlN. Journal of Physics and Chemistry of Solids, 48(7), 641-647.\u003c/li\u003e\n\u003cli\u003eHaider, S. T., Shah, M. A., Lee, D. G., \u0026amp; Hur, S. (2023). A review of the recent applications of aluminum nitride-based piezoelectric devices. IEEE Access, 11, 58779-58795.\u003c/li\u003e\n\u003cli\u003eOhta, J., Fujioka, H., Ito, S., \u0026amp; Oshima, M. (2002). Room-temperature epitaxial growth of AlN films. Applied Physics Letters, 81(13), 2373-2375.\u003c/li\u003e\n\u003cli\u003eLiu, B., Gao, J., Wu, K. M., \u0026amp; Liu, C. (2009). Preparation and rapid thermal annealing of AlN thin films grown by molecular beam epitaxy. Solid State Communications, 149(17-18), 715-717.\u003c/li\u003e\n\u003cli\u003eKhan, S., et al. (2015). Influence of the nitrogen fraction on AlN thin film deposited by cathodic arc ion. Materials Science in Semiconductor Processing, 29, 193-200.\u003c/li\u003e\n\u003cli\u003eBaba, S., Gajewski, W., Jasinski, M., Zelechowski, M., \u0026amp; Kazmierkowski, M. P. (2021). High performance power supplies for plasma materials processing. IEEE Access, 9, 19327-19344.\u003c/li\u003e\n\u003cli\u003eCarter, D. (2008). Arc prevention in magnetron sputtering processes. 51st Annual Technical Conference Proceedings of the Society of Vacuum Coaters, 380.\u003c/li\u003e\n\u003cli\u003eAnders, A. (2008). Cathodic arcs: From fractal spots to energetic condensation (Vol. 50). Springer.\u003c/li\u003e\n\u003cli\u003eAnders, A. (2006). Physics of arcing, and implications to sputter deposition. Thin Solid Films, 502(1-2), 22-28.\u003c/li\u003e\n\u003cli\u003eBogatin, E. (2004). Signal integrity: Simplified. Prentice Hall Professional.\u003c/li\u003e\n\u003cli\u003eBelkind, A., Freilich, A., \u0026amp; Scholl, R. (1998). Pulse duration effects in pulse-power reactive sputtering of Al2O3. Surface and Coatings Technology, 108, 558-563.\u003c/li\u003e\n\u003cli\u003eScholl, R. A. (1997). Power systems for reactive sputtering of insulating films. Surface and Coatings Technology, 93(1), 7-13.\u003c/li\u003e\n\u003cli\u003eManiv, S., \u0026amp; Westwood, W. D. (1980). Discharge characteristics for magnetron sputtering of Al in Ar and Ar/O2 mixtures. Journal of Vacuum Science and Technology, 17(3), 743-751.\u003c/li\u003e\n\u003cli\u003eCarter, D., Walde, H., \u0026amp; Nauman, K. (2012). Managing arcs in large area sputtering applications. Thin Solid Films, 520(12), 4199-4202.\u003c/li\u003e\n\u003cli\u003eSamukawa, S. (1996). Pulse‐time‐modulated electron cyclotron resonance plasma etching with low radio‐frequency substrate bias. Applied Physics Letters, 68(3), 316-318.\u003c/li\u003e\n\u003cli\u003eCaughman, J. B. O., \u0026amp; Hoffman, D. J. (1994). ARC detection and protection in high‐power antenna systems. AIP Conference Proceedings, 289(1), 379-382.\u003c/li\u003e\n\u003cli\u003eAmbrosy, U. (2012). TruPlasma RF series 3000, 7000 new. Proceedings of the 3rd International Conference on Power Electronics Plasma Engineering, 14.1-14.5.\u003c/li\u003e\n\u003cli\u003eWiedemuth, P., Ambrosy, U., Mann, E., Grede, A., \u0026amp; Thome, C. (2013). A new RF power supply platform for plasma applications. Proceedings of the 56th Annual Technical Conference of the Society of Vacuum Coaters, 571.\u003c/li\u003e\n\u003cli\u003ePankratz, J., \u0026amp; Hoffman, J. (2022). Understanding arc energy: Methods to measure, compare, and control arc energy. SVC Bulletin, Summer, 2-7.\u003c/li\u003e\n\u003cli\u003ePanjan, P., Drnov\u0026scaron;ek, A., Gselman, P., Čekada, M., \u0026amp; Panjan, M. (2020). Review of growth defects in thin films prepared by PVD techniques. Coatings, 10(5), 447.\u003c/li\u003e\n\u003cli\u003eVan Zyl, G., \u0026amp; Heckman, R. (2005). Managing arcs in RF powered plasma processes. Proceedings of the 48th Annual Technical Conference of the Society of Vacuum Coaters, 44-49.\u003c/li\u003e\n\u003cli\u003eCho, C., Kim, S., Choi, M., Lee, Y., Seong, I., Jeong, W., ... \u0026amp; You, S. J. (2024). Initiation mechanism of arcing generated in RF capacitively coupled plasma. Physica Scripta, 99(5), 055603.\u003c/li\u003e\n\u003cli\u003eHeintze, M., \u0026amp; Luciu, I. (2018). Nodule formation on sputtering targets: Causes and their control by MF power supplies. Surface and Coatings Technology, 336, 80-83.\u003c/li\u003e\n\u003cli\u003eOskirko, V. O., Semenov, V. D., Solovyev, A. A., Rabotkin, S. V., Pavlov, A. P., \u0026amp; Zakharov, A. N. (2022). Arc energy minimization in high-power impulse magnetron sputtering. Vacuum, 202, 111213.\u003c/li\u003e\n\u003cli\u003eGuljakow, J., \u0026amp; Lang, W. (2024). Analysis of arcing in processes during the reactive deposition of aluminum in a High Power Impulse Magnetron Sputtering process. hal-04669476f.\u003c/li\u003e\n\u003cli\u003eTiron, V., Velicu, I. L., Matei, T., Cristea, D., Cunha, L., \u0026amp; Stoian, G. (2020). Ultra-short pulse HiPIMS: A strategy to suppress arcing during reactive deposition of SiO2 thin films with enhanced mechanical and optical properties. Coatings, 10(7), 633.\u003c/li\u003e\n\u003cli\u003eWiedemuth, P., Merte, R., Richter, U., \u0026amp; Bannwarth, M. (2012). Next generation of mid-frequency power supplies for plasma applications. Proceedings of the 55th SVC Annual Technical Conference, 195-198.\u003c/li\u003e\n\u003cli\u003eWalde, H. V., Christie, D. J., Kovacevic, S., \u0026amp; Tomasel, F. G. (2012). Advances in power supplies for high arc rate applications. Proceedings of the 55th SVC Annual Technical Conference, 191-193.\u003c/li\u003e\n\u003cli\u003eYamada, A., Arai, H., Obayashi, S., \u0026amp; Shoki, H. (2011). Evaluation for RFID tag antennas by antenna clearance based on power reflection coefficient. Proceedings of the 5th European Conference on Antennas and Propagation (EUCAP), 2997-3000.\u003c/li\u003e\n\u003cli\u003eJun, H. S. (2013). Statistical prediction of micro-arcing in radio-frequency plasmas for the manufacture of semiconductor devices. Japanese Journal of Applied Physics, 52(6R), 066202.\u003c/li\u003e\n\u003cli\u003eRogers, J. H., LaRue, P., Phelps, D. A., \u0026amp; Pinsker, R. I. (1995). RF arc detection using harmonic signals. Proceedings of the 16th International Symposium on Fusion Engineering, 1, 522-525.\u003c/li\u003e\n\u003cli\u003eYan, E., et al. (2009). Solving arcing issues in CVD processes. ECS Transactions, 18(1), 581.\u003c/li\u003e\n\u003cli\u003eHeintze, M., \u0026amp; Kroyer, T. (2016). Benefit of modern arc management in MF dual magnetron sputtering power supplies: Stable deposition process for a wide range of materials. Vakuum in Forschung und Praxis, 28(4), 26-29.\u003c/li\u003e\n\u003cli\u003eYin, Y., et al. (2007). Time dependent plasma properties during microarcing in radio frequency plasmas. Applied Physics Letters, 91(19), 191504.\u003c/li\u003e\n\u003cli\u003eBu, G., Ciplys, D., Shur, M., Schowalter, L. J., Schujman, S., \u0026amp; Gaska, R. (2006). Surface acoustic wave velocity in single-crystal AIN substrates. IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control, 53(1), 251-254.\u003c/li\u003e\n\u003cli\u003eSeller, J. (1998). Asymmetric bipolar pulsed DC: The enabling technology for reactive PVD. Surface and Coatings Technology, 98(1-3), 1245-1250.\u003c/li\u003e\n\u003cli\u003eG\u0026uuml;ttler, D., Gr\u0026ouml;tzschel, R., \u0026amp; M\u0026ouml;ller, W. (2007). Lateral variation of target poisoning during reactive magnetron sputtering. Applied Physics Letters, 90(26).\u003c/li\u003e\n\u003cli\u003eRuiz, E., Alvarez, S., \u0026amp; Alemany, P. (1994). Electronic structure and properties of AIN. Physical Review B, 49(11), 7115.\u003c/li\u003e\n\u003cli\u003eBelkind, A., Freilich, A., Lopez, J., Zhao, Z., Zhu, W., \u0026amp; Becker, K. (2005). Characterization of pulsed DC magnetron sputtering plasmas. New Journal of Physics, 7(1), 90.\u003c/li\u003e\n\u003cli\u003eCarter, D., Walde, H., McDonough, G., \u0026amp; Roche, G. (2002). Parameter optimization in pulsed DC reactive sputter deposition of aluminum oxide. Proceedings of the Annual Technical Conference-Society of Vacuum Coaters, 570-577.\u003c/li\u003e\n\u003cli\u003eSoltanpour, P. N., Jones, J. B., Jr., \u0026amp; Workman, S. M. (1982). Optical emission spectrometry. In A. L. Page (Ed.), Methods of Soil Analysis: Part 2: Chemical and Microbiological Properties (pp. 29\u0026ndash;65). ASA.\u003c/li\u003e\n\u003cli\u003eQayyum, A., Zeb, S., Naveed, M., Ghauri, S., Waheed, A., \u0026amp; Zakaullah, M. (2006). Plasma Devices Operations, 14, 61.\u003c/li\u003e\n\u003cli\u003eKau, L. H., Huang, H. J., Chang, H. E., Hsieh, Y. L., Fuh, Y. K., \u0026amp; Li, T. T. (2020). Micro Nano Letters, 15, 323.\u003c/li\u003e\n\u003cli\u003eZhou, W. Y., Zhang, W. Z., Wang, L. S., \u0026amp; Zhang, H. J. (2023). Materials, 16, 3015.\u003c/li\u003e\n\u003cli\u003eHong, S. J., May, G. S., \u0026amp; Park, D. C. (2003). IEEE Transactions on Semiconductor Manufacturing, 16, 598.\u003c/li\u003e\n\u003cli\u003eChen, W. L., Lin, C. Y., Huang, J. M., \u0026amp; Chen, T. C. (2023). International Journal of Advanced Manufacturing Technology, 127, 2955\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"the-international-journal-of-advanced-manufacturing-technology","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"jamt","sideBox":"Learn more about [The International Journal of Advanced Manufacturing Technology](https://www.springer.com/journal/170)","snPcode":"170","submissionUrl":"https://submission.nature.com/new-submission/170/3","title":"The International Journal of Advanced Manufacturing Technology","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Reactive magnetron sputtering, Optical emission spectroscopy, Radio frequency (RF) 13.56MHz, Aluminum nitride (AlN), Arc management, Big data analysis","lastPublishedDoi":"10.21203/rs.3.rs-6737015/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6737015/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThis study focuses on the obtaining quality of arc-managed aluminum nitride (AlN) films in a PVD plasma system using a 13.56 MHz RF power source. Key parameters such as reflection coefficient (Г), suppression time, and re-ignition time were studied with the aim of reducing arc interference, increasing film crystallinity, and improving surface morphology. The key innovation of this work lies in the implementation of an ultrafast arc management strategy, with suppression/re-ignition times as short as 2 \u0026micro;s/2 \u0026micro;s\u0026mdash;significantly shorter than those typically reported in the literature (50\u0026ndash;1000 \u0026micro;s). This ultrashort modulation intercepts arc events at their micro-discharge stage, effectively minimizing energy accumulation and preventing the formation of high-energy \"hard arcs.\" Consequently, this approach stabilizes plasma conditions and leads to significant improvements in film crystallinity, surface smoothness, and deposition rate uniformity, offering a new pathway toward high-performance AlN thin film fabrication in RF sputtering systems. In the plasma spectra obtained from in situ optical emission spectroscopy (OES), significantly higher spectral intensities were observed during radio frequency (RF) sputtering, particularly the emission feature associated with ionized nitrogen (N₂⁺) at approximately 390.93 nm. This suggests that RF plasma provides energy conditions favorable for the formation of high-quality aluminum nitride (AlN) films. Furthermore, we applied the principal component analysis (PCA) algorithm for big data analysis to reduce dimensionality and visualize the clustering patterns of OES data recorded during thin film deposition under three different suppression/re-ignition conditions: 2 \u0026micro;s/2 \u0026micro;s, 100 \u0026micro;s/50 \u0026micro;s, and 1000 \u0026micro;s/500 \u0026micro;s.\u003c/p\u003e","manuscriptTitle":"Optical Emission Spectroscopy and Ultrafast Arc Management Strategy on the Quality of AlN Thin Films Using a 13.56 MHz RF Generator","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-05-30 12:01:09","doi":"10.21203/rs.3.rs-6737015/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Minor Revisions Needed","date":"2025-06-30T09:04:48+00:00","index":"","fulltext":""},{"type":"reviewerAgreed","content":"","date":"2025-05-27T19:17:23+00:00","index":0,"fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-05-27T17:40:31+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-05-27T04:54:19+00:00","index":"","fulltext":""},{"type":"submitted","content":"The International Journal of Advanced Manufacturing Technology","date":"2025-05-26T01:54:36+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"the-international-journal-of-advanced-manufacturing-technology","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"jamt","sideBox":"Learn more about [The International Journal of Advanced Manufacturing Technology](https://www.springer.com/journal/170)","snPcode":"170","submissionUrl":"https://submission.nature.com/new-submission/170/3","title":"The International Journal of Advanced Manufacturing Technology","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"becab2c1-ca11-439a-82cc-ddee090a98a6","owner":[],"postedDate":"May 30th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2025-07-28T16:05:32+00:00","versionOfRecord":{"articleIdentity":"rs-6737015","link":"https://doi.org/10.1007/s00170-025-16119-0","journal":{"identity":"the-international-journal-of-advanced-manufacturing-technology","isVorOnly":false,"title":"The International Journal of Advanced Manufacturing Technology"},"publishedOn":"2025-07-24 15:57:40","publishedOnDateReadable":"July 24th, 2025"},"versionCreatedAt":"2025-05-30 12:01:09","video":"","vorDoi":"10.1007/s00170-025-16119-0","vorDoiUrl":"https://doi.org/10.1007/s00170-025-16119-0","workflowStages":[]},"version":"v1","identity":"rs-6737015","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6737015","identity":"rs-6737015","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}