Constantin Haefner

The digital transformation of production requires methods for integrating, storing, and operationalizing data across organizational boundaries, yet most existing approaches remain siloed and unidirectional, lacking a systematic loop from raw data to actionable knowledge and back. We introduce Data-to-Knowledge (D2K) and Knowledge-to-Data (K2D) pipelines as a universal production concept built on networks of Digital Shadows. The Data-to-Knowledge (D2K) pipeline is realized as a cross-organizational proof of concept that captures and semantically annotates robotic trajectory data from three independent research institutes and uses those data to train an inverse-dynamics foundation model for robot control. Centralized aggregation via an existing FAIR-compliant research data repository was chosen deliberately over federated alternatives to maximize semantic interoperability and reuse of shared infrastructure; federated and privacy-preserving extensions are identified as a promising future direction. Fine-tuning the cross-organizationally trained foundation model reduces training time by approximately 85% relative to end-to-end training from scratch, while achieving comparable accuracy on a standardized inverse-dynamics benchmark. These gains are attributable to the combination of cross-site data aggregation and transfer learning; isolating the contribution of semantic annotation alone remains a topic for future ablation work. The implementation demonstrates that semantically enriched, cross-organizational D2K pipelines can accelerate model development and reduce redundant data collection within a constrained but practically relevant class of robotics tasks. We further discuss limitations, governance challenges, and how these pipelines can contribute to a broader World Wide Lab for collaborative production research.

Laser-based processing of materials has progressed from traditional applications such as macroscale cutting and welding, to sophisticated techniques, including ultrafast micromachining, additive manufacturing, and surface engineering at micro- and nano-metric scales. Innovations in laser source technology, particularly the advent of high-power and ultrashort-pulse lasers, have expanded the range of processable materials, enabled a plethora of manufacturing applications, and propelled groundbreaking research in optics, photonics, electronics, and biomedical domains. In this article, we provide a concise, yet broad review of the congruent evolution of lasers and materials processing, and highlight seminal developments in the field over the years, combined with a critical assessment of the state of the art. Finally, we also provide an outlook on emerging needs, as well as a roadmap of anticipated developments in laser technologies and materials platforms over the next 50 years. Graphic Abstract

Ti-6Al-4V is difficult to machine via milling or turning; therefore, near-net-shape manufacturing offers significant advantages as it can reduce the amount of material to be subtracted down to the final dimension. Additive manufacturing can be used to accomplish near-net-shapes, but its low deposition rate makes it less attractive for industry. In the present study, we investigated how a rectangular laser spot with a top hat intensity distribution can be used to increase the deposition rate of Ti-6Al-4V for powder-based laser direct energy deposition. Deposited single tracks, layers, and solid bodies were built to study the effect of processing parameters on the microstructure evolution. Post-thermomechanical treatment was carried out at different temperatures and strain rates using a deformation dilatometer. Track geometry, bonding defects, and microstructure were analyzed. The high deposition rate was increased up to 5 kg/h, while the process remained stable. Global shielding from the surrounding air is mandatory to prevent oxidation since the deposited volume heats up quickly.

The digital transformation of production requires new methods of data integration and storage, as well as decision making and support systems that work vertically and horizontally throughout the development, production, and use cycle. In this paper, we propose Data-to-Knowledge (and Knowledge-to-Data) pipelines for production as a universal concept building on a network of Digital Shadows (a concept augmenting Digital Twins). We show a proof of concept that builds on and bridges existing infrastructure to 1) capture and semantically annotates trajectory data from multiple similar but independent robots in different organisations and use cases in a data lakehouse and 2) an independent process that dynamically queries matching data for training an inverse dynamic foundation model for robotic control. The article discusses the challenges and benefits of this approach and how Data-to-Knowledge pipelines contribute efficiency gains and industrial scalability in a World Wide Lab as a research outlook.

In the additive manufacturing of components using powder-based laser metal deposition (LMD), the heating of the volume during buildup is a decisive factor for process stability and contour accuracy. If the process parameters remain constant, this intrinsic heating leads to deviations in the deposited layer thickness during the process because of changes in the melt pool volume. This leads to contour deviations and potentially causes the process to fail if the process parameters are no longer within the suitable range. Particularly in the case of complex geometries, this previously required time-consuming process development for adapted process parameters and buildup strategies. This paper examines the potential of data-driven approaches to enhance the stability and precision of LMD processes. To this end, a machine learning (ML) model is employed to optimize laser power settings. The objective of the study is to reduce the thermally induced geometric deviations that often occur during the LMD process by utilizing experimental data. The methodology employs the use of the alloy Inconel 718, renowned for its high strength and temperature resistance, in conjunction with the utilization of computer numerical control machines equipped with laser and imaging systems. The ML model is trained to predict the optimal laser power required to obtain consistent melt pool properties. The results demonstrate that the ML approach is an effective means of reducing geometric deviations.

A key challenge toward future quantum internet technology is connecting quantum processors at metropolitan scale. Here, we report on heralded entanglement between two independently operated quantum network nodes separated by 10 kilometers. The two nodes hosting diamond spin qubits are linked with a midpoint station via 25 kilometers of deployed optical fiber. We minimize the effects of fiber photon loss by quantum frequency conversion of the qubit-native photons to the telecom L-band and by embedding the link in an extensible phase-stabilized architecture enabling the use of the loss-resilient single-click entangling protocol. By capitalizing on the full heralding capabilities of the network link in combination with real-time feedback logic on the long-lived qubits, we demonstrate the delivery of a predefined entangled state on the nodes irrespective of the heralding detection pattern. Addressing key scaling challenges and being compatible with different qubit systems, our architecture establishes a generic platform for exploring metropolitan-scale quantum networks.

We present a highly efficient low-noise quantum frequency converter from the visible range to telecom wavelengths, combining a pump laser at intermediate frequency resonantly enhanced in an actively stabilized cavity with a monocrystalline bulk crystal. A demonstrator for photons emitted by nitrogen-vacancy-center qubits achieves 43% external efficiency with a noise photon rate per wavelength (frequency) band of 2 s⁻¹/pm(17 s⁻¹/GHz) – reducing the noise by two orders of magnitude compared with current devices based on periodically poled crystals with waveguides. With its tunable output wavelength, this device enables the generation of indistinguishable telecom photons from different network nodes and is, as such, a crucial component for a future quantum internet based on optical fiber.

SPIE Digital Library Proceedings

In Laser Powder Bed Fusion (LPBF), there are high demands on the cleanliness of laser protective windows. However, process by products may deposit on the protective windows during processing and causing changes to the process boundary conditions. In this work, a methodology to assess the condition of LPBF protective windows using a line scan camera is presented. Image processing is used to detect the degree of contamination on the surfaces by means of thresholding. Moreover, the influence of contaminated protective windows on the LPBF process is investigated with beam caustic measurements, analysis of single tracks, and investigation of relative sample density.

The unique combination of an intense heavy ion beam accelerator and a high energy laser opens the possibility of exploring new physics taking advantage of the synergy of both facilities. A variety of new fields can be addressed with this combination in plasma physics, atomic physics, nuclear- and astro-physics as well as material research. In addition, using CPA-technology, laser pulses with a pulse power of up to a petawatt opens the door to explore the regime of fully relativistic plasmas. Therefore the Gesellschaft fuer Schwerionenforschung is augmenting the current high intensity upgrade of the heavy ion accelerator facility with the construction of PHELIX. Designed with two pulse-generating front ends and send to multiple experimental areas PHELIX will serve as a highly versatile laser system for various applications. In this report, we present the design of the laser system and some key experiments that can be performed with this combination for the first time.