Magnetron Sputtering Uniformity vs Rate — PatSnap Eureka
Coating Thickness Uniformity vs. Deposition Rate in High-Power Magnetron Sputtering
The racetrack erosion zone of planar magnetrons creates an intrinsic conflict between throughput and uniformity — one that optical filter production, with tolerances as tight as 0.01% of mean layer thickness, cannot afford to ignore. PatSnap Eureka maps the full patent and literature landscape so your engineering team can resolve it faster.
Why Uniformity and Rate Are Intrinsically Coupled
The fundamental tension between deposition rate and coating uniformity in magnetron sputtering originates in the geometry and power density of the erosion zone on the sputtering target. In planar magnetron systems, the magnetic field confines plasma to a ring-shaped "racetrack," causing inhomogeneous erosion that directly imprints onto the particle flux profile reaching the substrate. As established by Saint Petersburg Electrotechnical University "LETI" (2018), the Danilin model predicts that increasing the cone angle of the target sputtering profile causes preferential thickening directly beneath the cathode while thinning the center and edges of the substrate.
Raising discharge power to increase throughput exacerbates this effect because higher power densities deepen the racetrack non-uniformity. Compounding the problem, as shown by Bauman Moscow State Technical University (2015), the erosion zone evolves continuously as the target is consumed — making both simulation and real-time control of uniformity difficult in high-throughput production. This is why IP landscape analytics of this space reveals sustained innovation pressure from academic and industrial players alike.
Working gas pressure introduces a further complication documented by internationally recognized thin-film standards bodies. Higher pressures thermalize sputtered particles more strongly, broadening the angular distribution and improving uniformity, but also reducing mean free path, increasing scattering losses, and modifying film microstructure. Seoul National University (2021) showed that at 80 mTorr compared to 20 mTorr, the density of charged Ti nanoparticles incorporated into the growing film changed substantially — pressure-driven uniformity improvements come at the direct cost of altered film nanostructure, which is critical to refractive index uniformity in optical interference filters.
The target-to-substrate distance is a central lever: increasing this distance spreads the particle flux more evenly but reduces the deposition rate due to geometric dilution. For a 100 mm planar Ni₇₉Fe₁₆Mo₅ target in argon, a peak rate of approximately 280 nm·min⁻¹ was measured, with larger target-substrate distances improving lateral uniformity across radial positions up to 60 mm — but necessarily reducing flux density at the substrate. This is a direct, unavoidable tradeoff with no hardware-free solution.
Quantifying the Uniformity–Rate Tradeoff
Key metrics from peer-reviewed literature and patent data, analyzed via PatSnap Eureka across 60+ sources spanning 7 countries.
HiPIMS Power Density vs. Film Hardness (Al–Si–N)
Changing average target power density from 30 to 120 W/cm² produces non-linear hardness gains — but constrains rate optimization. Tomsk Polytechnic University, 2019.
Non-Uniformity Achieved by Geometric Strategy
Comparing non-uniformity levels across different hardware and process strategies documented in the patent and literature dataset.
Deposition Rate: DC Magnetron vs. HiPIMS
HiPIMS pulsed duty cycle directly limits time-averaged deposition rate versus DCMS, despite superior film density and optical quality. CERN, 2020; Bauman MSTU, 2021.
Multi-Source Strategy: Non-Uniformity Cancellation
Dual-source compensating geometry partially cancels individual source non-uniformity, achieving <5% for WSi detector films. Bauman MSTU, 2020; Hunan City University, 2022.
High-Power Impulse Magnetron Sputtering: Film Quality vs. Throughput
HiPIMS is the clearest embodiment of the uniformity-rate tradeoff in the contemporary literature — superior optical film quality at an inherent rate penalty.
Denser, Void-Free Films at Challenging Incidence Angles
HiPIMS with positive pulse (voltage inversion after the main negative pulse) produced denser, void-free Nb thin films even at challenging grazing incidence geometries on copper substrates — improving optical quality compared to negatively biased DCMS. The approach imposes significant constraints on achievable deposition rate per unit time.
Advantage: void-free at grazing incidenceAbsence of Point Defect Absorption Centers
HiPIMS-deposited Al–Si–N films differ fundamentally from DCMS films due to the absence of point defect-related absorption centers — a direct advantage for optical filter transmission quality. Changing average target power density from 30 to 120 W/cm² produced non-linear changes in hardness (22–29 GPa) and absorption center concentration (10¹⁸–10²⁰ cm⁻³).
Hardness: 22–29 GPa across power rangeNanocrystalline and Amorphous Coatings — With a Rate Penalty
Pulsed methods including HiPIMS create preconditions for nanocrystalline and amorphous coatings with superior properties, while acknowledging the rate penalty inherent to pulsed operation. The duty cycle of pulsed operation is the fundamental constraint on time-averaged deposition rate — a constraint with no straightforward engineering workaround.
Trade: superior structure vs. lower throughputPressure-Driven Uniformity Alters Film Nanostructure
At 80 mTorr compared to 20 mTorr, the density of charged Ti nanoparticles incorporated into the growing film changed substantially. Pressure-driven uniformity improvements come at the cost of altered film nanostructure — directly relevant to the refractive index uniformity required in optical interference filters used in photonics and life sciences applications.
80 mTorr vs. 20 mTorr: significant nanostructure shiftGeometric and Multi-Source Approaches to Recovering Throughput
The optical filter industry has developed masking, multi-source, planetary motion, and erosion zone engineering to reconcile uniformity with deposition rate.
| Strategy | Institution / Patent Holder | Key Result | Rate Impact | Year |
|---|---|---|---|---|
| Mask aperture control | Tongji University IPOE | Minimum deposition beam width of 2.79 mm using 3 mm diameter aperture; 2D thickness profile correction of high-precision mirrors | ↓ Reduces fraction of sputtered material reaching substrate | 2020 |
| Planetary motion (rotation + revolution) | Tongji University | Thickness distribution model for laterally graded multilayer X-ray collimators; controls spatial gradient without masking rate penalty | ↑ No masking rate loss | 2021 |
| Dual-source compensating geometry | Bauman Moscow State Technical University | Non-uniformity below 5%; meets >95% uniformity requirement for WSi single-photon detector films | ↑ Recovers rate without masking | 2020 |
| Triple-target co-sputtering | Hunan City University | Global coordinate transformation to evaluate analytical uniformity formulae for arbitrary target-substrate geometries; optimizes source placement | ↑ Maximizes aggregate rate across large substrate | 2022 |
| Double-T erosion zone geometry | Oerlikon Surface Solutions AG (patent, active) | Two parallel tracks with curved end loop sections; increases flux from end loops to improve radial uniformity at substrate plane | ↑ Hardware-level — no rate penalty of masking | 2023 |
Need to assess freedom-to-operate on these geometric strategies?
PatSnap Eureka identifies active vs. expired patents across all assignees in seconds — powered by AI-native IP analytics.
Statistical Power Compensation: The Industrial State of the Art
The most advanced engineering response to the uniformity-rate tradeoff uses closed-loop control to modulate power in real time — compensating for target aging without sacrificing throughput.
Applied Materials: Statistical Model-Based Power Setpoints
Statistical analysis is performed on deposition profile measurements from previously processed substrates. A model of the deposition profile as a function of power parameter is fitted, and a power setpoint for the next substrate is determined from the fitted model. This feedforward-with-feedback architecture compensates for target aging without reducing deposition rate — power is increased to compensate for erosion rather than rate reduced to maintain uniformity. Patents filed in Taiwan (2024) and China (2025), both pending.
Carl Zeiss SMT: In-Situ Layer Stress Monitoring
Layer stress is monitored in real time during deposition, and the time course of stress is used to determine layer thickness and coating rate simultaneously. This non-contact measurement scheme avoids the rate penalty of interrupting deposition for ex-situ measurement, enabling high-uniformity, high-rate deposition to proceed continuously. Specific to EUV reflective coatings — among the most demanding optical applications in the industry. Patent: Carl Zeiss SMT GmbH, 2019.
Unaxis Balzers: Phase-Locked Sputtering Rate Adjustment
For magnetron systems with rotating magnetic field patterns, sputtering rate adjustment is phase-locked to the cyclic motion of the magnetron field pattern via an adjustment system. For optical coating of projection display components, the specification states layer thickness distribution must deviate from the mean by no more than 1% across at least 1000 cm². For optical data transmission substrates, no more than 0.01% across 10 cm². These tolerances make uncontrolled rate-uniformity tradeoffs operationally unacceptable. Patent: Unaxis Balzers (CN), 2010.
KIT ANKA: Growth Velocity Proportional to DC Power
In-situ X-ray reflectivity confirmed that growth velocity is proportional to DC power, validating the use of power as a direct rate actuator in closed-loop control systems. This proportionality, combined with the sensitivity of uniformity to power distribution across the erosion zone, is the physical basis for all power-compensation uniformity control architectures. This foundational result from KIT underpins the Applied Materials and Carl Zeiss approaches. Confirmed 2015.
Who Is Driving the Engineering Frontier?
Bauman Moscow State Technical University is the most frequently appearing academic institution in the dataset, contributing studies on erosion zone influence on uniformity, dual-source modeling for ultra-thin films, energy efficiency analysis, heat flux modeling, and liquid-phase magnetron source design. Their systematic approach to mathematical modeling of sputtering geometry underpins much of the quantitative understanding of the tradeoff.
Fraunhofer-Gesellschaft holds multiple active and inactive patents specifically targeting optical precision layer deposition, including methods for producing layer systems with high uniformity and low absorption (2003, 2014, 2015). Their work is distinguished by direct application to optical filter production with specified tolerances — the most industrially relevant academic-adjacent body of work in this dataset. Explore how leading R&D organizations use IP intelligence to benchmark against institutions like Fraunhofer.
Tongji University (IPOE/MOE Key Lab) contributes both mask-based particle distribution control and planetary motion sputtering system modeling, bridging fundamental particle transport physics and practical optical multilayer fabrication for diffractometers and X-ray mirrors. Applied Materials contributes the most industrially mature power compensation control architecture, with patents in Taiwan (2024) and China (2025) — the state of the art in closed-loop uniformity management at high deposition rates. For developers integrating these datasets programmatically, PatSnap's open API enables direct access to patent intelligence at scale.
Oerlikon Surface Solutions AG holds an active European patent (2023) on a double-T erosion zone geometry that improves flux uniformity through hardware design — a complementary, target-engineering approach. Anhui Guangzhi Technology (China) contributes multiple patents (2023–2024) on systematic target-to-substrate distance adjustment methods using resistance uniformity feedback — a pragmatic, production-floor approach that accepts rate as fixed and optimizes geometry iteratively. The full competitive landscape across these players is mappable in real time via PatSnap's innovation intelligence platform. For advanced materials applications, PatSnap's materials science solution provides dedicated thin-film and PVD patent analytics.
Magnetron Sputtering Uniformity vs. Rate — Key Questions Answered
In planar magnetron systems, the magnetic field confines plasma to a ring-shaped racetrack, causing inhomogeneous erosion that directly imprints onto the particle flux profile reaching the substrate. Raising discharge power to increase rate exacerbates this effect because higher power densities deepen the racetrack non-uniformity, causing preferential thickening directly beneath the cathode while thinning the center and edges of the substrate.
Optical filter production demands the tightest thickness tolerances in the PVD industry. For optical data transmission substrates, deviations must be no more than 0.01% of mean layer thickness across a 10 cm² area. For display projection optics, the tolerance is up to 1% across at least 1000 cm². These tolerances make unmanaged rate-uniformity tradeoffs commercially unacceptable.
HiPIMS delivers peak power densities orders of magnitude above conventional DC magnetron sputtering in short pulses, generating a highly ionized flux that improves film density, reduces roughness, and enables more conformal deposition. HiPIMS-deposited films differ fundamentally from DCMS films due to the absence of point defect-related absorption centers — an advantage for optical filter transmission quality. However, the duty cycle of pulsed operation inherently limits the time-averaged deposition rate.
Increasing target-to-substrate distance spreads the particle flux more evenly but reduces the deposition rate due to geometric dilution. For a 100 mm planar Ni₇₉Fe₁₆Mo₅ target in argon, larger target-substrate distances improved the lateral uniformity of the thickness distribution across radial positions up to 60 mm, but necessarily reduced the flux density at the substrate — a direct, unavoidable tradeoff.
Statistical model-based power compensation is the leading industrial approach. Applied Materials' patents use statistical analysis on deposition profile measurements from previously processed substrates. A model of the deposition profile as a function of power parameter is fitted to these measurements, and a power setpoint for the next substrate is determined from the fitted model. This feedforward-with-feedback architecture allows the system to compensate for drift in the erosion zone as the target ages without reducing deposition rate.
Using two sources in a compensating geometry allows the spatial non-uniformity of each individual source to be partially cancelled, recovering rate without compromising uniformity. Mathematical modeling of the dual-source geometry, based on experimentally measured mass flow distributions from individual magnetrons, achieved non-uniformity below 5% — meeting the greater than 95% uniformity requirement for WSi single-photon detector films. Triple-target co-sputtering extends this approach further, enabling system designers to optimize source placement for uniformity while maximizing aggregate deposition rate across a large substrate area.
Still have questions? Let PatSnap Eureka search the patent and literature database for you.
Ask Eureka AI Your Sputtering QuestionResolve the Uniformity–Rate Tradeoff Faster with AI Patent Intelligence
Join 18,000+ innovators already using PatSnap Eureka to accelerate their R&D — from sputtering process optimization to freedom-to-operate analysis across 120+ countries.
References
- Influence of the planar cylindrical target erosion zone of magnetron sputtering on the uniformity of a thin-film coating — Bauman Moscow State Technical University, 2015
- Magnetron discharge sputtering for fabrication of nanogradient optical coatings — Scientific-Manufacturing Enterprise "Fotron-Auto Ltd.", 2015
- Distribution of coating thickness applied by magnetron sputtering — Saint Petersburg Electrotechnical University "LETI", 2018
- Providing of Ultra-Thin Film Thickness Uniformity by Magnetron Sputtering from Two Sources — Bauman Moscow State Technical University, 2020
- Controlling Film Thickness Distribution by Magnetron Sputtering with Rotation and Revolution — Tongji University IPOE, 2021
- Collimated Magnetron Sputter Deposition for Mirror Coatings — Danish National Space Center, 2008
- Magnetic thin film deposition with pulsed magnetron sputtering: deposition rate and film thickness distribution — Electrotechnical Institute Division of Electrotechnology and Materials Science, Wroclaw, 2016
- Improved film density for coatings at grazing angle of incidence in high power impulse magnetron sputtering with positive pulse — CERN, 2020
- Theoretical and Experimental Study of Particle Distribution from Magnetron Sputtering with Masks for Accurate Thickness Profile Control — Tongji University IPOE, 2020
- Study on the Deposition Uniformity of Triple-Target Magnetron Co-Sputtering System: Numerical Simulation and Experiment — Hunan City University, 2022
- Method and device for producing layer systems for optical precision elements — Fraunhofer-Gesellschaft zur Forderung der Angewandten Forschung E.V., 2003
- Process and device for the production of uniform layers on moving substrates and layers produced in this way — Fraunhofer-Gesellschaft zur Förderung der Angewandten Forschung E.V., 2015
- Process for influencing the layer thickness distribution on substrates and use of a device for carrying out the process — Fraunhofer-Gesellschaft zur Förderung der Angewandten Forschung E.V., 2014
- Power compensation in PVD chambers — Applied Materials (Taiwan), 2024
- PVD Chamber Power Compensation — Applied Materials (China), 2025
- Methods for determining layer thicknesses, coating rates and/or process stability during the deposition of a coating — Carl Zeiss SMT GmbH, 2019
- Magnetron sputtering source and coating system arrangement — Oerlikon Surface Solutions AG, 2023
- Method for processing magnetron sputter-coated substrates — Unaxis Balzers (China), 2010
- Monitoring the thin film formation during sputter deposition of vanadium carbide — KIT/ANKA Institut für Photonenforschung und Synchrotronstrahlung, 2015
- Structure, Mechanical and Optical Properties of Silicon-Rich Al–Si–N Films Prepared by High Power Impulse Magnetron Sputtering — Tomsk Polytechnic University, 2019
- Pulsed methods of thin film coatings deposition — Bauman Moscow State Technical University, 2021
- Effect of Pressure on the Film Deposition during RF Magnetron Sputtering Considering Charged Nanoparticles — Seoul National University, 2021
- Simulation and Optimization of Film Thickness Uniformity in Physical Vapor Deposition — University of the West of Scotland, 2018
- Research on Thin Film Thickness Uniformity for Deposition of Rectangular Planar Sputtering Target — Northeastern University, 2012
- Improving the quality of nanofilms produced by magnetron sputtering — Saint Petersburg Electrotechnical University "LETI", 2021
- WIPO — World Intellectual Property Organization
- Karlsruhe Institute of Technology (KIT)
All data and statistics on this page are sourced from the references above and from PatSnap's proprietary innovation intelligence platform. Patent status information (active/inactive/pending) reflects status at time of dataset compilation and may have changed.
PatSnap Eureka searches patents and research to answer instantly.