A method for gradient-doped pulse sputtering chrome on a photo mask

Abstract

This invention relates to the field of photomask manufacturing technology and discloses a method for chromium plating using gradient-doped pulsed sputtering photomasks. The method includes cleaning and drying a glass substrate; activating it with an argon-oxygen mixed gas and a radio frequency plasma source to expose active broken bonds on the surface; moving the substrate to a chromium-silicon alloy target; activating a pulsed DC sputtering power supply; and performing a nitrogen linear gradient gas inlet program to deposit a gradient-doped transition layer, constructing a transition band that slowly releases lattice deformation energy; switching to a pure chromium target in situ; co-sponging the pulsed DC and RF power supplies with phase-synchronized triggering modulation to deposit a pure chromium functional layer; bombarding the film surface with a low-energy RF gas using an argon-nitrogen mixed gas; and initiating an in-situ isothermal annealing program for furnace cooling; finally, continuous polishing to obtain the finished product. This invention alleviates lattice mismatch and thermal stress by constructing a compositionally continuous transition band and using dual-power modulated sputtering, thereby improving interfacial adhesion and coating density.

Description

Technical Field

[0001] This invention relates to the field of photomask manufacturing technology, specifically to a method for chromium plating using gradient doped pulse sputtering photomasks. Background Technology

[0002] The photomask is the core template in the photolithography process, and the precision, uniformity, and reliability of its chromium plating layer directly determine the minimum linewidth and final yield of downstream chips and display devices. As semiconductor manufacturing processes advance to smaller nodes and display technologies continue to upgrade, high-end manufacturing has placed even more stringent standards on the chromium plating process of photomasks.

[0003] In existing manufacturing processes, there is a significant mismatch in the coefficients of thermal expansion between the glass substrate and the pure chromium layer. Furthermore, the chemical inertness of the glass substrate surface results in low interfacial adhesion between the chromium layer and the substrate. During subsequent processes such as photolithography and etching, or when subjected to temperature fluctuations, stress concentration easily occurs at the interface, leading to coating peeling or internal microcracks. As the size of the photomask substrate continues to increase, the accumulation of internal stress caused by this lattice mismatch can directly lead to macroscopic warping deformation of the substrate.

[0004] Traditional DC sputtering processes for chromium plating on large substrates suffer from edge thickness attenuation, making it difficult to meet the exposure accuracy requirements for high-resolution patterns. Layers grown using a single-power-source sputtering mode are prone to microscopic pores and pinholes, resulting in insufficient overall density. This makes pure chromium layers susceptible to over-etching with hydrofluoric acid-based etching solutions, hindering the preservation of high-precision linewidth sidewall morphology. While some conventional processes attempt to introduce oxygen or nitrogen doping to improve corrosion resistance, this further degrades layer density and increases surface roughness, failing to balance uniformity and chemical stability.

[0005] Conventional photomask coating processes often require removing the substrate from the cavity during multi-step switching, disrupting the vacuum environment. This not only increases the risk of interlayer interface contamination by external impurities but also incurs additional time losses. Furthermore, the sputtering deposition rate of traditional single-power sources is relatively low, making it difficult to efficiently process multiple large-size photomasks in the same batch, resulting in limited production turnaround efficiency and high unit manufacturing costs. Therefore, there is an urgent need to develop an integrated photomask chromium plating process that can simultaneously address the combined issues of interface adhesion, thin film density, and mass production efficiency. Summary of the Invention

[0006] To address the shortcomings of existing technologies, this invention provides a gradient doping pulse sputtering photomask chromium plating method, which solves the problems of low interfacial adhesion, stress concentration, and warping deformation of large-size substrates caused by lattice and thermal expansion coefficient mismatch between glass substrates and pure chromium layers in existing photomask chromium plating processes.

[0007] To achieve the above objectives, the present invention provides the following technical solution: This invention provides a method for chromium plating using a gradient-doped pulse sputtering photomask, comprising the following steps: The glass substrate is wet-cleaned and dried, then sent into a sputtering chamber and introduced with a mixture of argon and oxygen. The radio frequency plasma source is turned on for activation to obtain a pre-treated substrate. The pretreated substrate is moved to the chromium-silicon alloy target, the pulsed DC sputtering power supply is turned on, nitrogen gas is introduced and a linear gradient gas inlet program is executed, so that the nitrogen flow rate is smoothly increased from the initial flow rate to the final flow rate, and a gradient doped transition layer is deposited on the surface of the pretreated substrate. The in-situ switch was made to a pure chromium target, pure argon gas was introduced, and the pulsed DC sputtering power supply and the radio frequency power supply were turned on for coordinated sputtering. During the coordinated sputtering, phase-synchronous triggering modulation was performed to deposit a pure chromium functional layer on the surface of the gradient doped transition layer. A mixture of argon and nitrogen gas is introduced, and the radio frequency power supply is turned on to perform low-energy bombardment on the surface of the pure chromium functional layer. The in-situ annealing program is started for isothermal annealing, and the annealed substrate is obtained by furnace cooling. The annealed substrate is removed from the sputtering chamber and continuously polished using a silica sol alkaline polishing solution. After cleaning and drying, the finished photomask original is obtained.

[0008] By adopting the above technical solution, the introduction of radio frequency plasma surface activation, gradient doping transition structure, and dual-power phase-step modulation sputtering process can significantly improve the internal stress of the film and enhance the interface adhesion. The specific process mechanism is reflected in the following continuous reaction and deposition process: In the initial surface activation stage, considering that the glass substrate surface is mainly composed of a relatively stable silicon-oxygen network, the introduced argon-oxygen mixed gas forms a plasma under the excitation of a radio frequency source. High-energy argon ions are responsible for physically bombarding the substrate surface to remove residual hydrocarbon adsorbates, while free oxygen radicals directly participate in the chemical reactions of surface molecules. This synergistic physical and chemical action breaks the original silicon-oxygen bonds, thereby exposing a large number of highly chemically active free bonds on the substrate surface. The presence of these active dangling bonds allows the subsequently deposited film to directly capture bonding sites, transforming the originally simple physical van der Waals force adhesion into a more robust chemical bond.

[0009] When the activated substrate is moved to the target site, the transition layer is constructed using the silicon element in the chromium-silicon alloy target itself, ensuring that the initial film composition maintains a certain compositional affinity with the underlying silicon-oxygen network. As sputtering progresses, the nitrogen flow rate is set to increase linearly, and the reactive gas combines with the sputtered metal atoms to generate chromium-silicon compounds with varying nitrogen contents. During this process, the film composition gradually evolves from silicon-rich and low-nitrogen to chromium-rich and high-nitrogen from bottom to top. This continuous change in composition effectively creates a physical buffer zone between the substrate and the upper film, preventing lattice dislocations caused by abrupt compositional changes at the interface, thus mitigating the microscopic deformation energy induced by the difference in thermal expansion coefficients between the substrate and the pure chromium layer.

[0010] During the deposition of the pure chromium functional layer, a synergistic sputtering mode combining pulsed DC and radio frequency (RF) power supplies in a co-cavity was introduced. High-density argon ions in the RF environment continuously bombard the growing film with low energy, causing secondary migration of chromium atoms deposited on the surface and filling internal pores, thereby improving the overall density of the film. Furthermore, to prevent field interference caused by the dual-power co-cavity system, phase-synchronized triggering modulation logic was used to coordinate the discharge cycles of both power supplies, avoiding plasma instability issues caused by alternating electromagnetic field fluctuations.

[0011] After the main deposition is completed, the process moves to the post-processing stage. Before annealing, an argon-nitrogen mixed gas is introduced and bombards the chromium layer surface under radio frequency excitation. Nitrogen plasma passivates surface grain boundary defects and forms a thin surface nitrided region. The subsequent isothermal annealing operation provides the necessary thermodynamic conditions for short-range atomic rearrangement within the film, releasing the residual stress that inevitably accumulates during deposition.

[0012] Preferably, the method for preparing the chromium-silicon alloy target for the chromium-silicon alloy target site includes the following steps: Chromium powder with a purity of not less than 99.99% and high-purity silicon powder are mixed at a silicon atom content of 5-10% to obtain a mixed powder; The mixed powder is loaded into a ball mill jar and subjected to three-dimensional high-energy ball milling for 12-24 hours at a ball-to-powder mass ratio of 8:1-12:1 to obtain homogenized powder. The homogenized powder is loaded into a graphite mold and placed in a vacuum hot-pressing sintering furnace, with the absolute pressure of the vacuum being 0.0001-0.001 Pa. Under unidirectional axial pressure of 20-30MPa, heat to 1200-1350℃ at a heating rate of 8-12℃ / min, and hold at the temperature and pressure for 2-4 hours. After being naturally cooled to room temperature in the furnace, the chromium-silicon alloy target material is obtained through machining, surface grinding and backing plate bonding.

[0013] By employing the above technical solutions and controlling the silicon atom ratio range, the aim is to ensure that the sputtered transition layer retains sufficient silicon to match the glass substrate, while preventing excessive silicon content from causing embrittlement of the transition layer itself. Combined with high-energy ball milling and vacuum hot-pressing sintering, the internal crystalline phase distribution of the target material can be made more uniform, thereby reducing the probability of microparticle flaking or abnormal arcing during subsequent pulse sputtering.

[0014] Preferably, the glass substrate is a synthetic quartz glass substrate or a borosilicate glass substrate. The wet cleaning and drying steps specifically include: injecting a 5-10% sodium hydroxide aqueous solution into the cleaning tank and ultrasonically cleaning for 10-20 minutes at an ultrasonic frequency of 35-45 kHz and a temperature of 20-30°C; rinsing with deionized water and then immersing the glass substrate in a 1-2% hydrofluoric acid aqueous solution for 5-8 minutes; ultrasonically rinsing in deionized water at an ultrasonic frequency of 55-65 kHz and a temperature of 25-35°C for 8-12 minutes; and finally, vacuum drying in a vacuum drying oven at 110-130°C for 20-40 minutes.

[0015] By adopting the above technical solution, in the initial wet cleaning process, sodium hydroxide solution combined with ultrasonic operation is mainly used for saponification and removal of surface grease contaminants. The subsequent hydrofluoric acid immersion causes slight corrosion to the glass surface, removing the processing-damaged layer and an extremely thin layer of oxide impurities, thereby exposing the internal silica substrate and providing an interference-free interface for the subsequent plasma reaction.

[0016] Preferably, in the activation step: the absolute pressure of the background vacuum is 0.0001-0.0005 Pa; the volume ratio of argon to oxygen in the argon-oxygen mixture is 8:1-10:1; the radio frequency power of the radio frequency plasma source is set to 800-1200 W; and the activation time is 10-20 min.

[0017] By adopting the above technical solution, the atmosphere ratio and radio frequency power during activation are coordinated with each other. On the one hand, sufficient argon ions provide physical bombardment energy to break silicon-oxygen bonds, and on the other hand, an appropriate amount of oxygen free radicals are used to maintain the chemical reactivity of the newly broken bonds, so as to avoid these broken bonds from closing again in a short time or being secondary contaminated by residual hydrocarbons.

[0018] Preferably, in the step of depositing a gradient-doped transition layer on the surface of the pretreated substrate: the sputtering voltage of the pulsed DC sputtering power supply is set to 400-600V, the pulse frequency to 25-35kHz, and the duty cycle to 45-55%; the initial flow rate is 3-5sccm, and the final flow rate is 25-35sccm; the film deposition rate is controlled to be 5-8nm / min, and the total deposition thickness is 100-200nm.

[0019] By adopting the above technical solution, the pulsed DC parameters set during the transition layer deposition are mainly used to clean the charge accumulated on the target surface. Limiting the gradual range of nitrogen flow rate and deposition rate is to ensure that the evolution of nitrogen content inside the transition layer does not have obvious step breaks, ensuring the continuity of the composition gradient, and the reserved thickness range is just enough to bear and buffer the physical tension brought by the upper pure chromium thick film.

[0020] Preferably, in the step of phase-synchronous triggering modulation: when the pulsed DC sputtering power supply is in the negative voltage sputtering start phase interval, the output power of the RF power supply is controlled to be maintained at the set base power; when the pulsed DC sputtering power supply enters the turn-off or positive reverse voltage cleaning phase interval, the output power of the RF power supply is controlled to be reduced to 20-30% of the set base power value to avoid the superposition interference of alternating electromagnetic field on the DC reverse voltage cleaning process.

[0021] By adopting the above technical solution, regarding the specific operating logic of phase-synchronous modulation, since the target material needs to clean the accumulated charge on its surface during the forward and reverse voltage phase of the pulsed DC, if the RF power supply still maintains full load output at this time, the alternating electromagnetic field it generates can easily interfere with the cleaning path of the DC electric field, leading to cleaning obstruction. By appropriately reducing the RF power at this specific phase, the basic plasma is maintained from extinguishing, and electromagnetic interference is eliminated, thereby improving the stability of sputtering and suppressing the generation of arcing defects.

[0022] Preferably, in the co-sputtering step: the pure argon gas flow rate is set to 50-80 sccm, the working gas pressure of the chamber is adjusted to 0.3-0.5 Pa, and the heating temperature is 150-200℃; the pulsed DC sputtering power supply is set to a sputtering voltage of 600-800V, a pulse frequency of 45-55kHz, and a duty cycle of 40-60%; the RF power supply is set to a base power of 500-1000W and a fixed frequency of 13-14MHz; and the deposition thickness of the pure chromium functional layer is 100-1000nm.

[0023] By adopting the above technical solution, adjusting the working gas pressure and applying substrate heating during synergistic sputtering, sputtered atoms reaching the substrate are endowed with higher migration capabilities. Under the combined excitation of dual power supplies, the ions in the cavity exhibit a higher ionization rate, making the deposited pure chromium grains more compactly squeezed together, thereby ensuring the overall light-shielding performance of the finished product.

[0024] Preferably, in the steps of the low-energy bombardment treatment and in-situ annealing procedure: the volume ratio of argon and nitrogen in the argon and nitrogen mixture is 3:1-5:1; the RF power supply is set to 500-700W; the low-energy bombardment treatment time is 8-12min; the isothermal annealing temperature is controlled to rise to 250-300℃ and held for annealing for 60-90min; after annealing, the cooling rate is controlled to be 3-6℃ / min and cooled to room temperature with the furnace.

[0025] By adopting the above technical solution, the temperature range used in the annealing stage is sufficient to induce appropriate relaxation in the amorphous or microcrystalline deposited layer, reducing the density of point defects within the layer. By intervening in and controlling the relatively gentle cooling process, the possibility of induced thermal stress due to excessively rapid cooling is blocked, allowing the final mechanical properties of the film to tend to stabilize.

[0026] Preferably, in the continuous polishing step: an alkaline polishing slurry with a pH value adjusted to 9-10 is added dropwise; the polishing pressure of the grinding disc is set to 0.8-1.2 psi, the spindle speed is set to 200-300 rpm, and the continuous polishing time is 10-15 min.

[0027] By employing the above technical solution, the polishing process utilizes the weakly alkaline properties of silica sol to generate a slight chemical reaction with the chromium-plated surface, which, combined with the mechanical friction of the polishing disc, smooths out any remaining micro-protrusions on the surface. This planarization process, achieved through the interaction of chemical and mechanical processes, improves the surface roughness of the final photomask.

[0028] Preferably, in the step of wet cleaning and drying the glass substrate, the glass substrate has a length of 300-1780 mm, a width of 300-1500 mm, and a thickness of 3-10 mm.

[0029] By adopting the above technical solution, the limitation of size characteristics reflects that when facing wide glass substrates of different specifications, this method can also balance the stress distribution over a large area through the gradient doping and dual power supply synergy mechanism, and avoid the risk of warping caused by the enlargement of substrate size.

[0030] This invention provides a gradient-doped pulse sputtering method for chromium plating using a photomask. It offers the following advantages: 1. This invention utilizes radio frequency plasma surface activation and deposition of a linearly gradient-doped chromium-silicon-nitrogen transition layer to construct a buffer structure with continuously varying composition between a glass substrate and a pure chromium functional layer. This structure alleviates stress concentration caused by lattice mismatch and thermal expansion between the two materials, improves the peel strength at the glass-chromium layer interface, and maintains structural integrity after undergoing high and low temperature thermal cycling, making it less prone to peeling or microcracks. Simultaneously, it suppresses warping deformation of large-size substrates. This stress relief mechanism is compatible with various basic glass substrates, adapting to the processing requirements of semiconductor masks and flat panel display masks in different scenarios.

[0031] 2. This invention employs pulsed DC and RF dual-power supply co-sponging, coupled with phase-synchronized triggering modulation to avoid alternating electromagnetic interference. This ensures that the pure chromium functional layer receives continuous and stable low-energy ion bombardment during growth. The ion bombardment promotes secondary migration of deposited atoms and fills internal micropores, thereby improving the overall density of the coating and reducing surface roughness. The dense and uniform film structure not only improves the control precision of thickness uniformity and linewidth miniaturization, meeting the technical standards of advanced process photomasks, but also endows the finished product with excellent acid etching selectivity, making it less prone to oxidation and discoloration in long-term humid and hot environments, and ensuring chemical stability in complex processes.

[0032] 3. This invention integrates substrate activation, gradient transition layer deposition, pure chromium functional layer sputtering, and in-situ bombardment annealing into a single sputtering chamber for continuous operation. This avoids interface contamination and time loss caused by vacuum breaking during the process. Under the synergistic excitation of dual power supplies, a high plasma ionization rate is maintained, resulting in a higher film deposition rate compared to traditional single-power-supply processes. This continuous in-situ coating process supports the simultaneous processing of multiple large-size photomasks, effectively reducing unit manufacturing costs while improving wafer turnaround efficiency, demonstrating high adaptability to industrial mass production. Attached Figure Description

[0033] Figure 1 This is a bar chart showing the plasma stability and ignition frequency test results during the sputtering process of this invention. Figure 1 (a) is a comparison diagram of the maximum radio frequency reflection power fluctuation during the pure chromium layer deposition process of various embodiments of the present invention and comparative examples. Figure 1 (b) is a comparison chart of the cumulative number of micro-arc sparks in the same period between the embodiments of the present invention and the comparative examples; Figure 2 This is a distribution characteristic diagram of the macroscopic residual stress and critical scratch load test results of the present invention, wherein, Figure 2 (a) is a comparison diagram of macroscopic residual stress between various embodiments and comparative examples of the present invention. Figure 2 (b) is a comparison diagram of the critical scratch loads of various embodiments of the present invention and comparative examples; Figure 3 This is a distribution characteristic diagram of the uniformity of large-size optical film thickness and optical density of the present invention, wherein, Figure 3 (a) is a comparison diagram of film thickness non-uniformity in various embodiments and comparative examples of the present invention. Figure 3 (b) is a comparison diagram of the average optical density of various embodiments and comparative examples of the present invention; Figure 4 This is a distribution characteristic diagram of the chemical corrosion resistance and abrasion resistance test of the photomask film of the present invention, wherein, Figure 4 (a) is a comparison diagram of film thickness loss during chemical cleaning in various embodiments and comparative examples of the present invention. Figure 4(b) is a comparison diagram of the cyclic friction and wear depth of various embodiments and comparative examples of the present invention. Detailed Implementation

[0034] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0035] Preparation Examples 1-3: Preparation Example 1: This preparation example provides a method for preparing a chromium-silicon alloy target with a silicon atom content of 5%, including the following steps: Chromium powder with a purity greater than or equal to 99.99% was mixed with high-purity silicon powder at a silicon atomic percentage of 5%. The mixed powder was placed in a ball mill jar and subjected to three-dimensional high-energy ball milling for 12 hours at a ball-to-powder mass ratio of 10:1. The uniformly mixed powder was then placed in a graphite mold and placed in a vacuum hot-pressing sintering furnace. The furnace was evacuated to an absolute pressure less than or equal to 0.001 Pa. Under a unidirectional axial pressure of 20 MPa, the temperature was raised to 1200 °C at a heating rate of 10 °C / min and held at constant temperature and pressure for 2 hours. After naturally cooling to room temperature in the furnace, the chromium-silicon alloy target material was obtained through machining, surface grinding, and backing plate bonding.

[0036] Preparation Example 2: This preparation example provides a method for preparing a chromium-silicon alloy target with a silicon atomic percentage of 8%, comprising the following steps: Chromium powder with a purity greater than or equal to 99.99% was mixed with high-purity silicon powder at a silicon atomic percentage of 8%. The mixed powder was placed in a ball mill jar and subjected to three-dimensional high-energy ball milling for 18 hours at a ball-to-powder mass ratio of 10:1. The uniformly mixed powder was then placed in a graphite mold and placed in a vacuum hot-pressing sintering furnace. The furnace was evacuated to an absolute pressure less than or equal to 0.001 Pa. Under a unidirectional axial pressure of 25 MPa, the temperature was raised to 1250 °C at a heating rate of 10 °C / min and held at constant temperature and pressure for 3 hours. After naturally cooling to room temperature in the furnace, the chromium-silicon alloy target material was obtained through machining, surface grinding, and backing plate bonding.

[0037] Preparation Example 3: This preparation example provides a method for preparing a chromium-silicon alloy target with a silicon atom content of 10%, including the following steps: Chromium powder with a purity greater than or equal to 99.99% was mixed with high-purity silicon powder at a silicon atomic percentage of 10%. The mixed powder was placed in a ball mill jar and subjected to three-dimensional high-energy ball milling for 24 hours at a ball-to-powder mass ratio of 10:1. The uniformly mixed powder was then placed in a graphite mold and placed in a vacuum hot-pressing sintering furnace. The furnace was evacuated to an absolute pressure less than or equal to 0.001 Pa. Under a unidirectional axial pressure of 30 MPa, the temperature was increased to 1350 °C at a heating rate of 10 °C / min and held at constant temperature and pressure for 4 hours. After naturally cooling to room temperature in the furnace, the chromium-silicon alloy target material was obtained through machining, surface grinding, and backing plate bonding.

[0038] Examples 1-5: Example 1: This embodiment provides a gradient-doped pulse sputtering photomask chromium plating method, including the following steps: (1) Substrate wet cleaning and plasma activation pretreatment: A synthetic quartz glass substrate with dimensions of 600mm×600mm and a thickness of 6mm was placed in a cleaning tank and injected with an 8% sodium hydroxide aqueous solution. The substrate was ultrasonically cleaned for 15 minutes at an ultrasonic frequency of 40kHz and a temperature of 25℃. After rinsing with deionized water, the substrate was immersed in a 1.5% hydrofluoric acid aqueous solution for 6 minutes. Then, it was ultrasonically rinsed in deionized water at an ultrasonic frequency of 60kHz and a temperature of 30℃ for 10 minutes. The substrate was then transferred to a vacuum drying oven and vacuum-baked at 120℃ for 30 minutes. The dried substrate was then transferred to a sputtering chamber and the background vacuum was evacuated to 0.0005Pa. A mixture of argon and oxygen with a volume ratio of 9:1 was introduced, the radio frequency plasma source was turned on, the radio frequency power was set to 1000W, and the processing time was 15 minutes.

[0039] (2) Reactive sputtering deposition of the gradient doped transition layer: Maintaining uninterrupted vacuum in the cavity, the substrate was moved to the chromium-silicon alloy target obtained in Preparation Example 2. The pulsed DC sputtering power supply was turned on, and the sputtering voltage was set to 500V, the pulse frequency to 30kHz, and the duty cycle to 50%. Nitrogen gas was introduced, and a linear gradient gas inlet program was executed through the mass flow controller. The nitrogen flow rate was linearly and smoothly increased from the initial 5sccm to the final 30sccm over time. The film deposition rate was controlled at 6nm / min, and the total deposition thickness was 150nm.

[0040] (3) Pulse and RF co-sputtering of pure chromium functional layer based on phase-synchronous triggering: In-situ switching to a high-purity chromium target with a purity of 99.99%. Pure argon gas was introduced, with a flow rate set to 65 sccm, and the working pressure of the cavity was adjusted to 0.4 Pa. The substrate heating system was set to a temperature of 180℃. The pulse DC and RF dual-power co-cavity co-sputtering mode was activated. The pulse DC power supply was set to a sputtering voltage of 700V, a pulse frequency of 50kHz, and a duty cycle of 50%. The RF power supply was set to a base power of 800W and a fixed frequency of 13.56MHz. Phase-synchronous triggering modulation was executed through the hardware controller. When the pulse DC was in the negative pressure sputtering activation phase interval, the RF power supply output power was maintained at 800W. When the pulse DC entered the shutdown or positive reverse pressure cleaning phase interval, the RF power supply output power was instantly reduced to 200W (i.e., 25% of the set value). A pure chromium layer with a thickness of 800nm ​​was deposited.

[0041] (4) In-situ plasma post-treatment and vacuum annealing: Turn off the sputtering power supply and argon gas inlet valve to maintain the cavity's base vacuum. Introduce a mixture of argon and nitrogen gas with a volume ratio of 4:1. Turn on the RF power supply and set the power to 600W to perform low-energy plasma bombardment treatment on the deposited coating surface for 10 minutes. Start the in-situ annealing program in the cavity, raise the substrate area temperature to 280℃, and hold it at that temperature for 75 minutes. After annealing, control the cooling rate to 5℃ / min and cool it to room temperature with the furnace.

[0042] (5) Chemical mechanical polishing: The cooled substrate is removed from the vacuum chamber and loaded into a chemical mechanical polishing device. A silica sol alkaline polishing solution with a pH value adjusted to 9.5 is continuously added. The polishing pressure of the grinding disc is set to 1.0 psi and the spindle speed is set to 250 rpm. The polishing process is carried out continuously for 12 minutes. After cleaning and drying, the finished photomask original is obtained.

[0043] Example 2: This embodiment provides a gradient-doped pulse sputtering photomask chromium plating method, including the following steps: (1) Substrate wet cleaning and plasma activation pretreatment: A synthetic quartz glass substrate with dimensions of 300mm×300mm and a thickness of 3mm was placed in a cleaning tank and injected with a 5% sodium hydroxide aqueous solution. The substrate was ultrasonically cleaned for 15 minutes at an ultrasonic frequency of 40kHz and a temperature of 25℃. After rinsing with deionized water, the substrate was immersed in a 1% hydrofluoric acid aqueous solution for 5 minutes. Then, it was ultrasonically rinsed in deionized water at an ultrasonic frequency of 60kHz and a temperature of 30℃ for 10 minutes. The substrate was then transferred to a vacuum drying oven and vacuum-baked at 120℃ for 30 minutes. The dried substrate was then transferred to a sputtering chamber and the background vacuum was evacuated to 0.0005Pa. A mixture of argon and oxygen with a volume ratio of 9:1 was introduced, the radio frequency plasma source was turned on, the radio frequency power was set to 800W, and the processing time was 10 minutes.

[0044] (2) Reactive sputtering deposition of gradient doped transition layer: Maintaining uninterrupted vacuum in the cavity, the substrate was moved to the chromium-silicon alloy target obtained in Preparation Example 1. The pulsed DC sputtering power supply was turned on, and the sputtering voltage was set to 400V, the pulse frequency to 30kHz, and the duty cycle to 50%. Nitrogen gas was introduced as the reactive gas, and a linear gradient gas inlet program was executed through the mass flow controller. The nitrogen flow rate was linearly and smoothly increased from the initial 5sccm to the final 30sccm over time. The film deposition rate was controlled at 5nm / min, and the total deposition thickness was 100nm.

[0045] (3) Pulse and RF co-sputtering of pure chromium functional layer based on phase-synchronous triggering: In-situ switching to a high-purity chromium target with a purity of 99.99%. Pure argon gas was introduced, with a flow rate set to 50 sccm, and the working pressure of the cavity was adjusted to 0.3 Pa. The substrate heating system was set to a temperature of 150 °C. The pulse DC and RF dual-power co-cavity co-sputtering mode was activated. The pulse DC power supply was set to a sputtering voltage of 600 V, a pulse frequency of 50 kHz, and a duty cycle of 40%. The RF power supply was set to a base power of 500 W and a fixed frequency of 13.56 MHz. Phase-synchronous triggering modulation was executed through the hardware controller. When the pulse DC was in the negative pressure sputtering activation phase interval, the RF power supply output power was maintained at 500 W. When the pulse DC entered the shutdown or positive reverse pressure cleaning phase interval, the RF power supply output power was instantly reduced to 100 W (i.e., 20% of the set value). A pure chromium layer with a thickness of 100 nm was deposited.

[0046] (4) In-situ plasma post-treatment and vacuum annealing: Turn off the sputtering power supply and argon gas inlet valve to maintain the cavity's base vacuum. Introduce a mixture of argon and nitrogen gas with a volume ratio of 4:1. Turn on the RF power supply and set the power to 600W to perform low-energy plasma bombardment treatment on the deposited coating surface for 10 minutes. Start the in-situ annealing program in the cavity, raise the substrate area temperature to 250℃, and hold it at that temperature for 60 minutes. After annealing, control the cooling rate to 5℃ / min and cool it to room temperature with the furnace.

[0047] (5) Chemical mechanical polishing: The cooled substrate is removed from the vacuum chamber and loaded into a chemical mechanical polishing device. A silica sol alkaline polishing solution with a pH value adjusted to 9 is continuously added. The polishing pressure of the grinding disc is set to 0.8 psi and the spindle speed is set to 200 rpm. The polishing process is carried out continuously for 10 minutes. After cleaning and drying, the finished photomask original is obtained.

[0048] Example 3: This embodiment provides a gradient-doped pulse sputtering photomask chromium plating method, including the following steps: (1) Substrate wet cleaning and plasma activation pretreatment: A synthetic quartz glass substrate with dimensions of 1780mm×1500mm and a thickness of 10mm was placed in a cleaning tank and injected with a 10% sodium hydroxide aqueous solution. The substrate was ultrasonically cleaned for 15 minutes at an ultrasonic frequency of 40kHz and a temperature of 25℃. After rinsing with deionized water, the substrate was immersed in a 2% hydrofluoric acid aqueous solution for 8 minutes. Then, it was ultrasonically rinsed in deionized water at an ultrasonic frequency of 60kHz and a temperature of 30℃ for 10 minutes. The substrate was then transferred to a vacuum drying oven and vacuum-baked at 120℃ for 30 minutes. The dried substrate was then transferred to a sputtering chamber and the background vacuum was evacuated to 0.0005Pa. A mixture of argon and oxygen with a volume ratio of 9:1 was introduced, the radio frequency plasma source was turned on, the radio frequency power was set to 1200W, and the processing time was 20 minutes.

[0049] (2) Reactive sputtering deposition of gradient doped transition layer: Maintaining uninterrupted vacuum in the cavity, the substrate was moved to the chromium-silicon alloy target obtained in Preparation Example 3. The pulsed DC sputtering power supply was turned on, and the sputtering voltage was set to 600V, the pulse frequency to 30kHz, and the duty cycle to 50%. Nitrogen gas was introduced as the reactive gas, and a linear gradient gas inlet program was executed through the mass flow controller. The nitrogen flow rate was linearly and smoothly increased from the initial 5sccm to the final 30sccm over time. The film deposition rate was controlled at 8nm / min, and the total deposition thickness was 200nm.

[0050] (3) Pulse and RF co-sputtering of pure chromium functional layer based on phase-synchronous triggering: In-situ switching to a high-purity chromium target with a purity of 99.99%. Pure argon gas was introduced, with a flow rate set to 80 sccm, and the working pressure of the cavity was adjusted to 0.5 Pa. The substrate heating system was set to a temperature of 200℃. The pulse DC and RF dual-power co-cavity co-sputtering mode was activated. The pulse DC power supply was set to a sputtering voltage of 800V, a pulse frequency of 50kHz, and a duty cycle of 60%. The RF power supply was set to a base power of 1000W and a fixed frequency of 13.56MHz. Phase-synchronous triggering modulation was executed through the hardware controller. When the pulse DC was in the negative pressure sputtering activation phase interval, the RF power supply output power was maintained at 1000W. When the pulse DC entered the shutdown or positive reverse pressure cleaning phase interval, the RF power supply output power was instantly reduced to 300W (i.e., 30% of the set value). A pure chromium layer with a thickness of 1000nm was deposited.

[0051] (4) In-situ plasma post-treatment and vacuum annealing: Turn off the sputtering power supply and argon gas inlet valve to maintain the cavity's base vacuum. Introduce a mixture of argon and nitrogen gas with a volume ratio of 4:1. Turn on the RF power supply and set the power to 600W to perform low-energy plasma bombardment treatment on the deposited coating surface for 10 minutes. Start the in-situ annealing program in the cavity, raise the substrate area temperature to 300℃, and hold it at that temperature for 90 minutes. After annealing, control the cooling rate to 5℃ / min and cool it to room temperature with the furnace.

[0052] (5) Chemical mechanical polishing: The cooled substrate is removed from the vacuum chamber and loaded into a chemical mechanical polishing device. A silica sol alkaline polishing solution with a pH value adjusted to 10 is continuously added. The polishing pressure of the grinding disc is set to 1.2 psi and the spindle speed is set to 300 rpm. The polishing process is carried out continuously for 15 minutes. After cleaning and drying, the finished photomask original is obtained.

[0053] Example 4: This embodiment provides a gradient-doped pulse sputtering photomask chromium plating method, including the following steps: (1) Substrate wet cleaning and plasma activation pretreatment: A synthetic quartz glass substrate with a size of 1500mm×1200mm and a thickness of 8mm was placed in a cleaning tank, and its cleaning and plasma activation parameters were the same as those in Example 1.

[0054] (2) Reactive sputtering deposition of gradient doped transition layer: The target material and deposition parameters used in this step are the same as in Example 1.

[0055] (3) Pulse and RF co-sputtering of a pure chromium functional layer based on phase-synchronous triggering: In-situ switching to a high-purity chromium target with a purity of 99.99%. Pure argon gas was introduced, with a flow rate set to 70 sccm, and the working pressure of the cavity was adjusted to 0.4 Pa. The substrate heating system was set to 180 °C. The pulsed DC and RF dual-power co-cavity co-sputtering mode was activated and phase-synchronous triggering modulation was executed. The power, frequency, duty cycle, and RF power down-adjustment logic of the dual power supply were the same as in Example 1. A pure chromium layer with a thickness of 600 nm was deposited.

[0056] (4) In-situ plasma post-treatment and vacuum annealing: The post-treatment parameters and annealing parameters for this step are the same as those in Example 1.

[0057] (5) Chemical mechanical polishing: The polishing solution parameters and polishing conditions in this step are the same as in Example 1. After cleaning and drying, the finished photomask original is obtained.

[0058] Example 5: This embodiment provides a gradient-doped pulse sputtering photomask chromium plating method, including the following steps: A borosilicate glass substrate with dimensions of 600mm × 600mm and a thickness of 6mm was used instead of the synthetic quartz glass substrate in Example 1. All other steps and parameters, including wet cleaning and plasma activation pretreatment of the substrate, reactive sputtering deposition of the gradient doped transition layer, pulse and RF co-sputtering of the pure chromium functional layer based on phase-synchronized triggering, in-situ plasma post-treatment and vacuum annealing, and chemical mechanical polishing, were exactly the same as in Example 1. After cleaning and drying, the finished photomask master was obtained.

[0059] Comparative Examples 1-5: Comparative Example 1: Compared with Example 1, the difference is that the reactive sputtering deposition of the gradient doped transition layer in step (2) is omitted, and in step (3), only a pure chromium layer with a thickness of 950 nm is deposited using a single pulse DC sputtering power supply. The RF power supply is not turned on, and the phase-synchronous triggering modulation is not performed. All other aspects are the same.

[0060] Comparative Example 2: Compared with Example 1, the difference is that step (2) of reactive sputtering deposition of gradient doped transition layer is omitted, and the substrate after plasma activation pretreatment is directly deposited with step (3) pure chromium functional layer (the total deposition thickness is adjusted to 950nm to ensure consistent total film thickness), and the rest are the same.

[0061] Comparative Example 3: Compared with Example 1, the difference is that when nitrogen gas is introduced in step (2), the linear gradient gas introduction procedure is not executed. Instead, the nitrogen flow rate is set to a constant 17.5 sccm (i.e. the average flow rate at the beginning and end in Example 1) throughout the process. All other aspects are the same.

[0062] Comparative Example 4: Compared with Example 1, the difference is that although the pulse DC and RF dual power supply co-cavity co-sputtering mode is enabled in step (3), the phase-synchronous triggering modulation is not executed through the hardware controller, and the output power of the RF power supply is kept constant at 800W throughout the process (without power reduction with the phase interval of the pulse DC). The rest are the same.

[0063] Comparative Example 5: Compared with Example 1, the difference is that the operation of introducing mixed gas and turning on the radio frequency plasma source for activation is omitted in step (1) (only wet cleaning and vacuum baking are performed), and the in-situ plasma post-treatment and vacuum annealing operation in step (4) are completely omitted, and chemical mechanical polishing is performed directly. The rest are the same.

[0064] Test Examples 1-4: Test Example 1: This test case is to verify the feasibility of the physical mechanism of the dual-power cooperative sputtering mechanism based on phase-synchronous triggering proposed in this invention in suppressing electromagnetic crosstalk, eliminating abnormal discharge on the target surface, and maintaining the overall stability of the plasma.

[0065] The magnetron sputtering systems used in the preparation processes of Examples 1 to 5 and Comparative Examples 1 to 5 were selected as monitoring objects.

[0066] During the pure chromium functional layer deposition stage, a high-frequency online power meter is connected to the RF matching network of the sputtering system, while a high-frequency oscilloscope and a micro-arc counting module are connected in parallel at the output of the pulsed DC power supply. Comparative Example 1, because the RF power supply is not turned on, only the DC data is monitored.

[0067] A unified monitoring time window was set to cover the entire actual physical deposition cycle of the pure chromium layer, and the maximum fluctuation difference of the radio frequency reflection power was fully recorded by an online power meter.

[0068] The cumulative number of micro-arc sparks captured by the pulsed DC terminal during the entire deposition cycle was counted, and the experimental data were recorded and exported.

[0069] Table 1. Plasma stability test data for examples and comparative examples in conclusion: According to the data in Table 1, the maximum RF reflection power fluctuation in Examples 1 to 5 remained below 5.1W throughout the entire deposition cycle, and the cumulative number of micro-arc sparks was low. Actual process observation revealed that this was directly due to the introduction of a phase-synchronous triggering modulation mechanism. (Refer to...) Figure 1 The extremely low power fluctuation data shown in embodiment (a) confirms that when the pulsed DC is in the sensitive phase range of the forward reverse voltage cleaning process, the underlying hardware controller forces the output power of the RF power supply to be instantly reduced, causing a millisecond-level response decrease in the background plasma concentration within the cavity, thereby effectively avoiding the superimposed interference of high-energy electromagnetic fields on the DC reverse voltage cleaning process. (Refer to...) Figure 1 (b) Examples 2 and 5 achieved an extremely ideal zero-arc state. A careful review of their experimental settings reveals that Example 2 employed a relatively low initial sputtering voltage and RF power configuration, resulting in a decrease in the transient power density on the target surface, thereby weakening the driving force for charge accumulation in the insulating micro-regions. Example 5, while maintaining the parameters of Example 1, introduced a borosilicate glass substrate. The difference in dielectric constant and thermal conductivity brought about by the different materials fine-tuned the macroscopic impedance network within the sputtering chamber. Combined with dual-power supply peak-shifting modulation, a plasma steady-state equilibrium region with minimal impedance fluctuations was established, thereby suppressing the occurrence of abnormal discharge on the target surface.

[0070] In contrast, Comparative Example 4, which employed dual-power co-cavity sputtering while maintaining a constant RF power output, exhibited RF reflected power fluctuations increasing to 87.4W, accompanied by severe target surface arcing occurring up to 156 times. The continuous high-frequency RF field and the reverse voltage of the pulsed DC created intense localized plasma oscillations near the target surface, leading to the instantaneous high-energy breakdown of the insulating micro-regions. This intense target surface discharge directly ejects molten metal droplets onto the substrate surface below, forming particle defects that cannot be repaired by cleaning later.

[0071] Observing the data of Comparative Example 1, its maximum RF reflected power fluctuation was 0.0W. This does not mean that the RF matching network has reached a perfect lossless state, but rather that the comparative example was set to completely disable the RF power supply during the pure chromium functional layer deposition stage. In this pure DC working environment without RF assistance, even though the pulse power supply itself has a reverse voltage cleaning function, after long-term continuous sputtering, the residual micro-charges on the target surface are still difficult to be completely neutralized by a single power supply, ultimately leading to 14 arcing events. Combined with the low fluctuation performance of the embodiment, the precise peak-shifting control of the dual-power supply collaborative mechanism in the time domain effectively blocked the physical causes of abnormal discharge while maintaining a high target ionization rate. Although Comparative Examples 2 and 3 are also equipped with synchronous trigger control, Comparative Example 2 lacks a bottom transition layer, while Comparative Example 3 lacks a gradient doping design. The mismatch in the micro-physical properties of the interface between the two caused a slight local plasma impedance change in the early stage of deposition, resulting in a slightly higher number of arcing events than the conventional embodiment. This indirectly confirms the deep parametric coupling relationship between the film structure design and the bottom power supply modulation.

[0072] Test Example 2: This test case is to verify the feasibility of the underlying physical metallurgical mechanism of the present invention, which involves introducing a gradient doped transition layer between the glass substrate and the pure chromium functional layer and implementing specific in-situ activation and annealing treatments, for releasing macroscopic residual stress in the system and improving the physical bonding force of the interface.

[0073] Test samples from the preparation stages of Examples 1 to 5 and Comparative Examples 1 to 5 were selected as the test objects.

[0074] A thin-film curvature measuring instrument based on the principle of laser reflection is used to perform a two-dimensional full-area laser scan on the surface of the substrate before plasma activation pretreatment, and the original macroscopic curvature radius is extracted as the reference data.

[0075] After all the coating, annealing and polishing processes are completed for each group of samples, they are placed back into the curvature measuring instrument for a second scan to obtain the curvature radius after coating. The difference between the two curvature values ​​is converted into the macroscopic residual stress value of the entire film according to the Stoney equation.

[0076] After the above tests were completed, the sample was transferred to the test platform of the automatic scratch tester. Using a standard diamond indenter with a radius of curvature of 200 micrometers, a vertical load increasing linearly from 1N to 100N was applied to the coating surface, with the scratch length set to 10 mm and the scratching speed to 5 mm / min.

[0077] Real-time acquisition of acoustic emission signal intensity and friction coefficient abrupt change data during the scratching process; the instantaneous vertical load corresponding to the first continuous strong peak of acoustic emission signal is recorded as the critical scratch load for film peeling.

[0078] Table 2. Test data of residual stress and bonding strength of the examples and comparative examples in conclusion: Based on the data in Table 2, combined with Figure 2 (a) and Figure 2 (b) The intuitive distribution characteristics show that the macroscopic residual stress of the entire photomask system was stably controlled at a low level below 220 MPa in the example group, while exhibiting a critical scratch load of approximately 50 N. Pure chromium metal has a densely packed body-centered cubic lattice, which easily accumulates high compressive stress during vacuum deposition. Furthermore, the significant mismatch in thermal expansion coefficients between the metal and the amorphous glass substrate makes internal stress during the cooling stage a long-standing industrial problem plaguing production lines. (Refer to...) Figure 2 In (a), the data points of Comparative Examples 1 and 2 show that, due to the direct deposition of a pure chromium layer on the glass surface, the underlying lattice buffer structure is missing, leading to an increase in macroscopic residual stress to 845.2 MPa and 756.8 MPa, respectively. This high-stress state directly weakens the adhesion of the film layer. Figure 2 (b) Comparative Examples 1 and 5, both groups, exhibited brittle spalling after withstanding a load of less than 20 N in the scratch test. The chromium-silicon alloy doped transition zone introduced in the examples is rich in silicon atoms near the glass side, enabling it to form good bonding compatibility with the silicon-oxygen network of the substrate. Subsequent nitrogen gradient introduction gradually induces the formation of amorphous or microcrystalline nitride phases with high strain tolerance. This series of operations constructs a physical transition zone capable of moderately releasing lattice deformation energy, effectively cutting off the path of rapid stress transmission from the bottom layer to the functional layer.

[0079] Simply changing the composition is insufficient to maintain system equilibrium; gradient control constitutes the core closed loop of this mechanism. Observation Figure 2 (a) and Figure 2(b) Comparative Example 3 shows that when nitrogen gas is introduced at a constant flow rate, although the residual stress decreases compared to the case without a transition layer and the critical load increases to 36.2 N, there is still a significant difference compared to the example. Constant doping creates a compositional abrupt interface between the transition layer and the pure chromium layer. This interface not only blocks the inter-layer diffusion of atoms but also forms a new stress concentration surface due to the discontinuity of the lattice constants on both sides. The linear gradient gas flow rate control results in a smooth transition of the chemical composition of the interface, eliminating a clear macroscopic interface and thus dispersing the concentrated stress throughout the thickness of the transition layer.

[0080] Reference Figure 2 The load data in Comparative Example 5 (b) demonstrates the limitations of relying solely on vapor deposition. After omitting the oxygen plasma activation and vacuum annealing steps, the adhesion strength of this group drops to a minimum of 15.2 N. The glass surface without oxygen plasma bombardment cannot expose a sufficient number of active silanol bonds, and the sputtered metal atoms can only adhere to the substrate with weak van der Waals forces, lacking strong chemical bonding. The missing in-situ vacuum annealing step further results in the permanent retention of microscopic defects such as vacancies and interstitial atoms generated during deposition within the film layer, preventing short-range atomic rearrangement and lattice relaxation through thermal activation. Considering the overall excellent performance of the embodiments, in-situ plasma activation after wet cleaning provides the basis for interfacial chemical bonding, while the continuous physical vapor deposition of the gradient transition layer, coupled with a specific annealing temperature profile, jointly contributes to a low-stress and high-strength film-substrate system. Test Example 3: This test case aims to verify the feasibility of the mechanism of the dual-power supply collaborative process of the present invention in eliminating large-size edge effects and improving optical consistency. The focus is on evaluating the macroscopic distribution characteristics of optical thin films in the effective process area of ​​each group of samples.

[0081] Large-size photomask prototypes prepared in Examples 1 to 5 and Comparative Examples 1 to 5 were selected as experimental subjects.

[0082] A fully automated thin film thickness gauge was used to set a 15×15 equally spaced grid within the effective exposure area of ​​the substrate. The physical thickness of the pure chromium functional layer was automatically scanned across the entire area, and the measured values ​​at each point were recorded. The film thickness range and film thickness non-uniformity percentage of the entire board were calculated.

[0083] The test sample was then transferred to the operating table of the transmission optical densitometer, positioned, and aligned with the same grid test coordinates as described above.

[0084] Using a 365nm wavelength ultraviolet light source, the vertical transmitted light intensity was measured point by point and its logarithmic value was converted into optical density (OD) value. The average optical density of the entire plate and the corresponding standard deviation of the data were then derived.

[0085] Table 3. Test data on film thickness uniformity and optical parameters of the examples and comparative examples in conclusion: Based on the data in Table 3, combined with Figure 3 (a) and Figure 3 (b) The macroscopic distribution characteristics show that the film thickness non-uniformity of the entire plate in the example group was generally suppressed to within 2.4%, and the average optical density of the entire plate consistently exceeded the design threshold of 3.0, demonstrating extremely high in-plane uniformity. Semiconductor photomask fabrication has long faced the problem of plasma density decay at large-size edges. The magnetron constraint force formed by the traditional single-pulse DC power supply at the target edge weakens sharply, leading to a sharp reduction in metal sputtering yield in the substrate edge region. (Refer to...) Figure 3 The low columnar distribution exhibited in embodiment (a) confirms that the radio frequency plasma source introduced in this scheme spatially disperses the electron distribution, and the reciprocating heating effect of the high-frequency alternating electromagnetic field on the electrons broadens the effective radial range of the high-density plasma. This spatial electric field broadening effect brought about by dual-power coupling physically compensates for the lack of deposition rate in the edge region, enabling the condensation rate of gas phase atoms to maintain a high degree of consistency on a substrate scale of several hundred square centimeters.

[0086] The spatial uniformity of physical thickness will inevitably be projected onto the macroscopic optical transmittance distribution. (Refer to...) Figure 3 In (a) of Comparative Example 1, the data bar shows that because the RF field assistance was stripped during the deposition stage, the edge film thickness attenuation caused by pure DC discharge worsened the film thickness non-uniformity to 6.54%. This directly resulted in this group having a lower film thickness uniformity. Figure 3 (b) shows that the average optical density scatter plot drops to a low level of around 2.85, and the data standard deviation is significantly amplified. Under the high-intensity light exposure of an actual lithography machine, regions with OD values ​​below 3.0 are prone to weak ultraviolet light penetration and leakage, leading to unexpected photosensitivity of the underlying photoresist. Further observation of the data trend in Comparative Example 3 reveals that a sudden interface induced by constant flow of nitrogen gas induces micropores caused by stress unevenness within the film layer. These micropores alter the refractive index of the bulk material, as shown in the reference diagram. Figure 3 (b) The corresponding data point has an optical density test value that hovers around 2.98, failing to achieve the ideal complete light-blocking density.

[0087] Reference Figure 3 (a) and Figure 3In Comparative Example 4 (b), the distribution state, maintaining a constant RF power output, resulted in severe target surface arc discharge. This not only disrupted the normal sputtered atom flight trajectory, causing the film thickness range to rise to 125.7 nm, but also created dense physical scattering centers on the film surface due to the large solid droplets sputtered by the arc. The ultraviolet detection beam emitted by the densitometer experienced strong diffuse reflection and scattering losses when passing through these rough and defective micro-regions in Comparative Example 4, leading to a complete disruption of the effective transmission signal received by the detector, and reducing its average optical density to an unusable level of 2.62. Considering the parameter performance of the embodiments, phase-synchronous trigger modulation is not only an electrical control method but also a core physical barrier for maintaining the macroscopic optical density and spatial geometric uniformity of the film.

[0088] Test Example 4: This test case is to verify the microstructure compactness and interfacial mechanical stability of the optical film system of the present invention under extreme chemical cleaning environment and high-frequency physical contact.

[0089] Standardized test samples prepared in Examples 1 to 5 and Comparative Examples 1 to 5 were used as the experimental subjects in this study, and the samples were divided into chemical corrosion test area and physical friction test area.

[0090] In the chemical corrosion test, a standard SPM cleaning solution (a mixture of concentrated sulfuric acid and 30% hydrogen peroxide in a volume ratio of 3:1) was prepared for semiconductor production lines. The solution was then heated and kept at a constant temperature of 90°C.

[0091] After accurately measuring the initial physical thickness of the functional layer of each sample using an elliptic polarization spectrometer, the sample was completely immersed in the above-mentioned high-temperature SPM solution for 30 minutes. After being taken out, it was ultrasonically rinsed with deionized water and dried with nitrogen. The film thickness was measured again, and the difference in film thickness loss before and after cleaning was recorded.

[0092] In the physical friction test, the sample was fixed on the platform of a reciprocating linear friction and wear testing machine. A standard PTFE friction ball head with a diameter of 5 mm was selected. Under a constant vertical load of 2.5 N, the reciprocating stroke was set to 15 mm and the sliding frequency to 2 Hz.

[0093] The test was conducted for 1000 cycles of reciprocating friction. After the test, a white light interferometer was used to scan the three-dimensional surface morphology of the middle section of the friction trajectory, extract the maximum wear depth of the wear track section, and summarize the experimental data.

[0094] Table 4. Test data on chemical corrosion resistance and abrasion resistance of the examples and comparative examples in conclusion: Based on the data in Table 4, combined with Figure 4 (a) and Figure 4 (b) The specific distribution trend of the example groups showed good structural stability in stringent physical and chemical destructive tests, with film thickness loss controlled within 1.5 nm and wear depth lower than conventional processes. In actual semiconductor photomask manufacturing and usage lifecycles, photomasks need to undergo multiple SPM (Surface Mount Perfection) removal cleanings and periodic physical brushing; the microstructure density of the thin film directly determines its resistance to corrosion. (Refer to...) Figure 4 The thickness loss data in Comparative Example 1 (a) confirms that the lack of a radio frequency plasma-assisted pure DC vapor deposition process prevents sputtered atoms from acquiring sufficient surface migration energy during the film formation stage. This low-energy-state-grown chromium film contains numerous columnar grain boundaries and micropores. A mixture of high-temperature sulfuric acid and hydrogen peroxide penetrates deep into the film along these grain boundary channels, causing rapid dissolution of the bulk material. The embodiment, by introducing a dual-power-supply synergistic phase-synchronization modulation mechanism, significantly enhances the atomic kinetic energy reaching the substrate surface through the continuous heating effect of the alternating radio frequency field on the space plasma. This promotes the transformation of the film growth from loose columnar crystals to dense equiaxed crystals or amorphous states, thereby sealing the microscopic pathways for the chemical etching solution to penetrate.

[0095] observe Figure 4 (b) In Comparative Example 4, the wear depth reached 24.58 nm. Frequent micro-arc sparking caused by maintaining constant RF power output left a large number of unmelted metal droplets and corresponding pits on the film surface. These macroscopic defects, under the reciprocating scraping of the PTFE ball head, easily become stress concentration fracture sources, leading to large-area brittle spalling of the surrounding film material. After eliminating abnormal discharge interference such as sparking, the stress buffer structure of the film-substrate interface became another core factor determining the wear resistance life. In Comparative Example 2, due to the direct deposition of pure chromium on the glass, the residual stress accumulated in the bottom layer was continuously excited during 1000 high-frequency physical friction cycles. The lattice distortion energy that was not effectively released at the interface superimposed with the external mechanical shear force, inducing the rapid nucleation and propagation of microcracks inside the film, ultimately manifesting as severe structural wear of 15.64 nm.

[0096] Referring to the high damage values ​​in both tests of Comparative Example 5, the negative impact of omitting plasma activation and vacuum annealing steps on the microscopic stability of the film is clearly demonstrated. The microscopic vacancies and dislocation networks that could not be eliminated through heat treatment exhibit higher chemical activity in a strong acid oxidizing environment, exacerbating the atomic dissolution rate. Simultaneously, the loose film-substrate interface lacking chemical bonding undergoes microscopic slippage under reciprocating mechanical stress, losing its overall resistance to deformation. Considering the low loss performance of the overall embodiment, the stress buffer transition layer constructed by linear gradient doping, and the high-density crystal structure resulting from precise peak-shifting modulation by dual power supplies, a highly reliable optical film system possessing both high chemical inertness and good mechanical toughness is achieved.

Claims

1. A method for chromium plating using a gradient-doped pulse sputtering photomask, characterized in that, Includes the following steps: The glass substrate is wet-cleaned and dried, then sent into a sputtering chamber and introduced with a mixture of argon and oxygen. The radio frequency plasma source is turned on for activation to obtain a pre-treated substrate. The pretreated substrate is moved to the chromium-silicon alloy target, the pulsed DC sputtering power supply is turned on, nitrogen gas is introduced and a linear gradient gas inlet program is executed, so that the nitrogen gas flow rate is smoothly increased from the initial flow rate to the final flow rate, and a gradient doped transition layer is deposited on the surface of the pretreated substrate. The in-situ switch is made to a pure chromium target, pure argon gas is introduced, the pulsed DC sputtering power supply and the radio frequency power supply are turned on to perform coordinated sputtering, and phase-step synchronous triggering modulation is performed in the coordinated sputtering to deposit a pure chromium functional layer on the surface of the gradient doped transition layer. A mixture of argon and nitrogen gas is introduced, and the radio frequency power supply is turned on to perform low-energy bombardment on the surface of the pure chromium functional layer. The in-situ annealing program is started for isothermal annealing, and the annealed substrate is obtained by furnace cooling. The annealed substrate is removed from the sputtering cavity, and the annealed substrate is continuously polished using a silica sol alkaline polishing solution. After cleaning and drying, the finished photomask original is obtained.

2. The gradient doping pulse sputtering chromium plating method according to claim 1, characterized in that, The method for preparing the chromium-silicon alloy target for the chromium-silicon alloy target site includes the following steps: Chromium powder with a purity of not less than 99.99% and high-purity silicon powder are mixed at a silicon atom content of 5-10% to obtain a mixed powder; The mixed powder is loaded into a ball mill jar and subjected to three-dimensional high-energy ball milling for 12-24 hours at a ball-to-powder mass ratio of 8:1-12:1 to obtain homogenized powder. The homogenized powder is loaded into a graphite mold and placed in a vacuum hot-pressing sintering furnace, with the absolute vacuum pressure being 0.0001-0.001 Pa. Under unidirectional axial pressure of 20-30MPa, heat to 1200-1350℃ at a heating rate of 8-12℃ / min, and hold at the temperature and pressure for 2-4 hours. After being naturally cooled to room temperature in the furnace, the chromium-silicon alloy target material is obtained through machining, surface grinding, and backing plate bonding.

3. The gradient doping pulse sputtering chromium plating method according to claim 1, characterized in that, The glass substrate is a synthetic quartz glass substrate or a borosilicate glass substrate, and the wet cleaning and drying steps specifically include: Inject a sodium hydroxide aqueous solution with a mass concentration of 5-10% into the cleaning tank, and perform ultrasonic cleaning for 10-20 minutes at an ultrasonic frequency of 35-45kHz and a temperature of 20-30℃. After rinsing with deionized water, the glass substrate is immersed in a 1-2% hydrofluoric acid aqueous solution and left to stand for 5-8 minutes. Transfer to deionized water with an ultrasonic frequency of 55-65kHz and a temperature of 25-35℃ for ultrasonic rinsing for 8-12 minutes; Transfer to a vacuum drying oven and bake at 110-130℃ for 20-40 minutes.

4. The gradient doping pulse sputtering chromium plating method according to claim 1, characterized in that, In the activation step: The absolute pressure of the background vacuum is 0.0001-0.0005 Pa; In the argon and oxygen mixture, the volume ratio of argon to oxygen is 8:1-10:1; The radio frequency power of the radio frequency plasma source is set to 800-1200W, and the activation time is set to 10-20min.

5. The gradient doping pulse sputtering chromium plating method according to claim 1, characterized in that, In the step of depositing a gradient-doped transition layer on the surface of the pretreated substrate: The pulsed DC sputtering power supply is set to a sputtering voltage of 400-600V, a pulse frequency of 25-35kHz, and a duty cycle of 45-55%. The initial flow rate is 3-5 sccm, and the final flow rate is 25-35 sccm; The film deposition rate was controlled at 5-8 nm / min, and the total deposition thickness was 100-200 nm.

6. The gradient doping pulse sputtering chromium plating method according to claim 1, characterized in that, In the step of phase-synchronous triggering modulation: When the pulsed DC sputtering power supply is in the negative voltage sputtering start phase interval, the output power of the radio frequency power supply is controlled to be maintained at the set base power; When the pulsed DC sputtering power supply enters the shutdown or forward reverse voltage cleaning phase range, the output power of the radio frequency power supply is controlled to be reduced to 20-30% of the set base power value to avoid superposition interference of alternating electromagnetic field on DC reverse voltage cleaning process.

7. The gradient doping pulse sputtering chromium plating method according to claim 6, characterized in that, In the steps of the coordinated sputtering: The pure argon gas flow rate is set to 50-80 sccm, the working gas pressure of the cavity is adjusted to 0.3-0.5 Pa, and the heating temperature is 150-200℃; The pulsed DC sputtering power supply is set with a sputtering voltage of 600-800V, a pulse frequency of 45-55kHz, and a duty cycle of 40-60%. The radio frequency power supply is set to a base power of 500-1000W and a fixed frequency of 13-14MHz; The deposition thickness of the pure chromium functional layer is 100-1000 nm.

8. The gradient doping pulse sputtering chromium plating method according to claim 1, characterized in that, In the steps of the low-energy bombardment treatment and the in-situ annealing procedure: In the argon and nitrogen mixture, the volume ratio of argon to nitrogen is 3:1 to 5:1; The power of the radio frequency power supply is set to 500-700W, and the low-energy bombardment treatment time is set to 8-12 minutes; The isothermal annealing temperature is raised to 250-300℃ and held for annealing for 60-90 minutes. After annealing, the cooling rate is controlled at 3-6℃ / min and the furnace is cooled to room temperature.

9. The gradient doping pulse sputtering chromium plating method according to claim 1, characterized in that, In the continuous polishing step: Add the silica sol alkaline polishing solution with the pH adjusted to 9-10; Set the polishing pressure of the grinding disc to 0.8-1.2 psi, the spindle speed to 200-300 rpm, and the continuous polishing time to 10-15 min.

10. The gradient doping pulse sputtering chromium plating method according to claim 1, characterized in that, In the steps of wet cleaning and drying the glass substrate: The glass substrate has a length of 300-1780mm, a width of 300-1500mm, and a thickness of 3-10mm.

Images