A revolution that has transformed the chemical, pharmaceutical and biomedical industries is now taking over thin-film production methods and vapour deposition tools. Traditionally, materials development has relied upon sequential experimentation and characterisation, where researchers methodically vary one parameter at a time. This −—–, while thorough, is inherently time-consuming and resource-intensive. In the pharmaceutical industry, high-throughput synthesis and screening methods have accelerated drug discovery and similar strategies are now being adopted in the thin-film industry. Combinatorial sputtering systems represent a paradigm shift in materials research, enabling scientists to explore vast compositional and processing spaces with unprecedented efficiency. When coupled with artificial intelligence and machine learning, these systems are revolutionising how we discover and optimise advanced materials.

What is Combinatorial Sputtering?

Combinatorial sputtering is a high-throughput physical vapour deposition technique that allows researchers to deposit thin films with systematically varied compositions, thicknesses, or processing parameters across a single substrate. By utilising multiple sputtering targets and carefully controlled deposition geometries, these systems can create material libraries containing hundreds of distinct compositions in a single experimental run.

Targets can be run with different powers and positioned at specific angles to create compositional gradients, and substrate manipulation can be programmed to achieve precise thickness variations. The result is a materials library on single wafers that would require months or years to produce using conventional methods, created instead in hours or days.

Key Advantages for Research Programmes

Extraordinary Exploration Efficiency

The most compelling advantage of combinatorial sputtering is the dramatic acceleration of the materials discovery process. Research teams can explore multi-dimensional compositional spaces that would be impractical to investigate using traditional methods. For instance, a ternary alloy system that might require 100 individual depositions using conventional sputtering can be comprehensively mapped in a single combinatorial experiment.

This efficiency extends beyond mere time savings. Combinatorial approaches reduce consumption of expensive target materials, minimise substrate costs, and decrease energy usage per data point generated. For research institutions operating under constrained budgets, this represents a significant economic advantage while simultaneously increasing scientific output.

Improved Data Quality and Consistency

When multiple compositions are deposited simultaneously on a single substrate, many experimental variables remain constant across the entire library. Substrate temperature, chamber pressure, background atmosphere, and deposition history are identical for all samples. This intrinsic consistency eliminates many sources of systematic error that plague sequential experiments, where chamber conditions may drift between runs or substrates may vary in quality.

The result is higher quality data that allows researchers to discern subtle compositional trends and identify optimal formulations with greater confidence. Comparative studies become more reliable when the primary variable is composition itself, rather than a combination of composition and uncontrolled experimental factors.

Enabling Discovery of Non-Intuitive Materials

Materials science has historically been guided by empirical rules and chemical intuition. However, many high-performance materials exhibit properties that emerge from complex interactions not easily predicted by simple heuristics. Combinatorial methods excel at revealing these non-intuitive compositions by systematically sampling regions of the vast materials space that researchers might otherwise overlook.

High-entropy alloys, which derive their properties from configurational entropy across five or more elements, exemplify materials that benefit from combinatorial exploration. Their vast compositional space and often unexpected property combinations make them ideal candidates for high-throughput discovery methods.

The AI-Enabled Revolution in Combinatorial Research

The integration of artificial intelligence and machine learning with combinatorial sputtering and high-throughput characterisation methods represents the forefront of autonomous materials discovery. These systems combine the material library data generation capabilities of combinatorial synthesis methods and suitable characterisation techniques with the pattern recognition and optimisation capabilities of AI, co-operating powerful methods for directing materials research.

Automated Characterisation and Data Management

Modern combinatorial systems generate enormous datasets. A single substrate library might contain characterisation data from hundreds of locations, including composition, structure, mechanical properties, optical characteristics, and electrical behaviour. Managing and extracting meaningful insights from this data volume represents a significant challenge.

Machine learning algorithms excel at handling such high-dimensional datasets. Automated image analysis can rapidly assess microstructural features from scanning electron microscopy. Classification algorithms can identify phase boundaries and structural transitions. Regression models can interpolate properties across compositional gradients, providing continuous rather than discrete mapping of the phase space.

Active Learning and Experimental Design

Perhaps the most transformative aspect of AI integration is the implementation of active learning loops. Rather than simply analysing data after experiments are complete, AI systems can guide the experimental process itself. These systems employ sophisticated algorithms that identify which experiments will be most informative given current knowledge, propose optimal next experiments, and iteratively refine understanding.

Bayesian optimisation, a particularly powerful approach, builds probabilistic models of the relationship between process conditions, material composition and properties. The algorithm balances exploration of poorly characterised regions with exploitation of promising areas, efficiently navigating toward optimal deposition conditions and material compositions. This approach can reduce the number of experiments required to identify high-performance materials by an order of magnitude compared to systematic grid searches.

Property Prediction and Virtual Screening

Advanced machine learning models trained on combinatorial data can predict material properties for compositions not yet synthesised. These predictive capabilities enable virtual screening of large compositional spaces, with physical experiments focused on the most promising candidates. Techniques such as neural networks, Gaussian process regression, and ensemble methods have all proven effective for materials property prediction.

The accuracy of these predictions improves as more data accumulates, creating a propitious cycle where each experiment enhances the model’s predictive power. Over time, these systems develop sophisticated understanding of composition-property relationships within specific material families.

Driving Applications Across Research Fields

Functional Coatings and Thin Films

Combinatorial sputtering has proven particularly valuable for developing functional coatings where multiple competing objectives must be optimised. Transparent conducting oxides, for example, require simultaneous optimisation of electrical conductivity and optical transparency. Combinatorial libraries allow researchers to map the trade-off between these properties across the probed process–structure–properties space and identify optimal formulations.

Magnetic materials for recording media, thermoelectric materials for energy conversion, and phase-change materials for memory applications have all benefited from combinatorial approaches. The technique is especially powerful for multicomponent systems where traditional trial-and-error methods become prohibitively time-consuming.

Catalytic Materials

Heterogeneous catalysts represent another application domain where combinatorial methods provide substantial advantages. Catalyst performance depends critically on surface composition, electronic structure, and morphology, all of which can be systematically varied using combinatorial sputtering. High-throughput screening of catalytic activity, combined with composition mapping, enables rapid identification of promising formulations.

The ability to create compositional gradients across support materials is particularly valuable, allowing researchers to optimise both the active catalyst phase and its interaction with the support in a single experiment.

Protective and Wear-Resistant Coatings

Industrial applications often demand coatings that combine multiple properties: hardness, wear resistance, corrosion protection, and thermal stability. Combinatorial methods enable systematic exploration of compositional effects on these properties, facilitating the development of next-generation protective coatings.

Nitride and carbide systems, including titanium aluminium nitride and chromium-based hard coatings, have been extensively studied using combinatorial approaches. The technique allows researchers to map hardness, oxidation resistance, and tribological performance as functions of composition, identifying optimal formulations for specific applications.

Integration with Complementary Techniques

Combinatorial sputtering systems deliver maximum value when integrated with high-throughput characterisation methods. X-ray diffraction mapping reveals phase evolution across compositional gradients. Automated nanoindentation assesses mechanical properties at multiple library locations. Optical spectroscopy characterises electronic and optical behaviour. Electrical probe stations evaluate charge transport properties.

This integration requires careful planning of substrate layouts and characterisation strategies. Successful implementations often incorporate specially designed substrate holders that facilitate multiple characterisation techniques while maintaining position registration. Data management systems must track measurements from numerous analytical tools and correlate them with compositional information.

Practical Considerations for Research Laboratories

System Design and Configuration

Effective combinatorial sputtering systems require careful consideration of target configuration, substrate manipulation capabilities, and process monitoring. Confocal target arrangements, where multiple sources are positioned around a central substrate, are common for creating compositional spreads. Rotating substrates, moving shutters, and programmable power supplies provide the flexibility needed for various library designs.

Modern systems increasingly incorporate in-situ monitoring capabilities, including optical emission spectroscopy for process control and quartz crystal microbalances for deposition rate measurement. These features enhance reproducibility and enable real-time adjustments during library creation.

Data Infrastructure

The data-intensive nature of combinatorial research demands robust data management infrastructure. Databases must accommodate diverse data types from multiple characterisation techniques while maintaining clear linkages to sample position, process conditions and composition. Open-source frameworks like the Materials Data Facility, NOMAD and the Digital Materials Foundry have emerged to support experimental materials data management, which are complemented by a growing number of computational materials databases.

Integration with computational tools for data analysis, visualisation, and machine learning requires standardised data formats and well-documented workflows. The investment in data infrastructure pays dividends through improved research efficiency and enhanced ability to extract insights from accumulated data.

Future Directions and Emerging Trends

The Materials 4.0 revolution, powered by AI and big data, continues to evolve rapidly. Emerging trends include the integration of multiple deposition techniques on single platforms, enabling exploration of processing parameter spaces alongside compositional variation. Korvus Technology is proud to offer highly modular deposition systems, integrating an unrivalled selection of sources and instruments that provide researchers with the flexibility needed to succeed in this latest phase of materials discovery. Learn more about our selection of sputtering and evaporation sources on our website as well as other innovative third-party instruments integrated onto our HEX platform. Alternatively, get in contact with a technical sales representative for further information.

The convergence of combinatorial synthesis, high-throughput characterisation, and artificial intelligence represents a fundamental transformation in how materials research is conducted. For institutions investing in these capabilities, the benefits extend beyond immediate research productivity to include training of students in data-driven experimental science and positioning at the forefront of modern materials discovery.

Conclusion

Combinatorial sputtering systems have matured from specialised research tools to essential infrastructure for competitive materials research programmes. Their ability to rapidly explore compositional process–structure–properties spaces, coupled with the pattern recognition capabilities of machine learning, enables discovery of advanced materials at unprecedented rates. As AI integration becomes more sophisticated and autonomous experimentation systems more capable, the pace of materials innovation will continue to accelerate.

For research institutions committed to advancing materials science, investment in combinatorial capabilities represents not merely an incremental improvement but a strategic transformation in research approach. The question is no longer whether to adopt these methods but how quickly they can be implemented and integrated into existing research programmes.

References

Dudarev, V., Banko, L. & Ludwig, A. An extensible open-source solution for research digitalisation in materials science. npj Computational Materials 11, 116 (2025). https://doi.org/10.1038/s41524-025-01618-1

Green, M. L., Choi, C. L., Hattrick-Simpers, J. R., Joshi, A. M., Takeuchi, I., Barron, S. C., Campo, E., Chiang, T., Empedocles, S., Gregoire, J. M.& Zakutayev, A. (2017). Fulfilling the promise of the materials genome initiative with high-throughput experimental methodologies. Applied Physics Reviews, 4(1), 011105 (2017). https://doi.org/10.1063/1.4977487

Henry Royce Institute, & Halden Miller, S. (2025). National Materials Innovation Strategy. Zenodo. https://doi.org/10.5281/zenodo.14603871

Kusne, A. G., Yu, H., Wu, C., Zhang, H., Hattrick-Simpers, J., DeCost, B., Sarker, S., Oses, C., Toher, C., Curtarolo, S. & Davydov, A. V. On-the-fly closed-loop materials discovery via Bayesian active learning. Nature Communications 11, 5966 (2020). https://doi.org/10.1038/s41467-020-19597-w

Ludwig, A. Discovery of new materials using combinatorial synthesis and high-throughput characterization of thin-film materials libraries combined with computational methods. npj Computational Materials 5, 70 (2019). https://doi.org/10.1038/s41524-019-0205-0

Stein, H. S., & Gregoire, J. M. Progress and prospects for accelerating materials science with automated and autonomous workflows. Chemical Science 10(42), 9640–9649 (2019). https://doi.org/10.1039/C9SC03766G

Szymanski, N. J., Zeng, Y., Huo, H., Bartel, C. J., Kim, H., & Ceder, G. Toward autonomous design and synthesis of novel inorganic materials. Materials Horizons, 8(8), 2169–2198 (2021). https://doi.org/10.1039/D1MH00495F

Warren, J. A. The Materials Genome Initiative and artificial intelligence. MRS Bulletin 43, 452–457 (2018). https://doi.org/10.1557/mrs.2018.122

Xiang, X. D., Sun, X., Briceno, G., Lou, Y., Wang, K. A., Chang, H., Wallace-Freedman, W. G., Chen, S. W. & Schultz, P. G. A combinatorial approach to materials discovery. Science, 268(5218), 1738–1740 (1995). https://doi.org/10.1126/science.268.5218.1738

Zakutayev, A., Wunder, N., Schwarting, M., Perkins, J. D., White, R., Munch, K., Tumas, W., & Phillips, C. An open experimental database for exploring inorganic materials. Scientific Data, 5, 180053 (2018). https://doi.org/10.1038/sdata.2018.53