1 Foundational Technology Layers

Physical and functional integration (PFI) considers the development of new elements and methods enabling more functionalities to be physically integrated on CMS in the most effective form factor. This requires interdisciplinary technology innovations as smart CMS may utilize a combination of features based on nano-electronics, micro-electro-mechanic, thermoelectric, magnetic, photonic, micro-fluidic, optical, acoustic, radiation, radio frequency, biological, chemical and quantum principles. Furthermore, many types of devices are to be integrated together, such as sensors, actuators, energy generators, energy storage devices, data processing devices, transceivers and antennae. Different technological approaches such as mainstream silicon technologies, MEMS/NEMS, MOEMS and LAE (Large Area Electronics) can be combined for the synergistic assembly of electronic and photonic devices.

PFI goes beyond the compact monolithic SoC approaches supported by the semiconductor technologies covered in the Process Technology, Equipment, Materials and Manufacturing Chapter. It addresses System in Package (SiP) realizations involving different integration methods of components with a higher degree of technological heterogeneity into different platforms, including Integrated Circuit carriers, PCB boards, additive manufacturing on flexible or hybrid substrates, and co-packaging of optics and electronics.

PFI requires not only the integration of physical components together, but also the co-design and integration of software, especially embedded software and embedded AI accelerators, to create reliable and sustainable functional systems. This also extends to the development of modular architectures that enable such systems to be configured dynamically and optimally for a given application. However, this chapter concentrates on the physical integration of hardware, including computational devices into systems, while leaving the software edge to the Embedded Software Chapter.

Given the broad range of physical scenarios they face, smart CMS need to interact with many environments, once outside the lab, including demanding ones such as harsh environments or in-vivo.

CMS face numerous operational challenges related to energy, performance, and size. In terms of energy, portable IoT devices require ultra-low-power operation and some form of energy autonomy—either through self-powering or by providing sufficient autonomy to meet specific application requirements (e.g., lifetime, service intervals). At the same time, for power electronics applications, they must address issues of high current density and the resulting thermal stress. For size, achieving the optimal minimum footprint at a given application is critical. This calls for the use of heterogeneous integration alongside advanced packaging and interconnection technologies to deliver maximum performance in the smallest possible system-level form factor, with minimal power consumption—or at least with consideration of available energy resources. In high-end consumer electronics, high-performance, ultra-dense, and compact interfaces across all integration levels—from chiplets and chip assembly/packaging to component and module connectors – shall be met. Increasingly, these systems may also incorporate photonics-based communication to meet data transfer and efficiency needs.

The development of MEMS/NEMS/MOEMS technologies centers on sensors and actuators that exploit the free surfaces and volumes achievable through semiconductor-based microfabrication. Sensors include next-generation inertial measurement units (accelerometers and gyroscopes) with enhanced performance—optionally augmented by AI—alongside magnetometers, pressure sensors, microphones, and particle sensors. The integration of advanced spintronic sensors, particularly those based on TMR technology, will further enhance these capabilities by providing ultra-sensitive, low-power magnetic field detection suitable for edge AI. This will enable Europe to secure technological sovereignty in a strategic, high-value sensing domain. Actuators encompass piezoelectric, electrostatic, or electromagnetic devices such as micro-mirrors, print heads, oscillators (membranes and cantilevers), tunable lenses, loudspeakers, and micromachined ultrasound transducers (pMUTs). Material innovations are also significant. For example, scandium aluminum nitride (ScAlN) enables new capabilities and improved performance in piezoelectric MEMS, supporting applications such as acoustic RF filters, optical systems like LiDARs, and ultrasonic sensing. This miniaturization is particularly critical for medical and consumer health applications, where MEMS enable the development of advanced sensors for miniaturized diagnostic equipment. Such technologies are foundational for enabling devices at the point of care, reducing the complexity of diagnostics and the time to results. For consumer health, this trend facilitates the integration of sensors into everyday devices for continuous monitoring of biomarkers. Integration compatibility remains a key consideration—for instance, thermal imagers (e.g., MOEMS microbolometer arrays) can be integrated as MOEMS layers above IC components.

In addition, silicon-based terahertz (THz) components and systems are emerging as a critical enabler for next-generation smart CMS. Leveraging advances in CMOS and BiCMOS technologies, highly integrated THz transceivers, sensors, and imaging modules are becoming feasible, offering compact, low-cost, and energy-efficient solutions. These technologies unlock new modalities for high-resolution imaging, spectroscopy, ultra-high-capacity wireless communications, and advanced radar sensing, complementing MEMS, photonics, and quantum systems. The expected outcome is a pathway toward scalable, miniaturized THz systems-on-chip that can be co-integrated with photonic and electronic subsystems, enabling multifunctional, high-performance smart modules for applications ranging from healthcare diagnostics and industrial inspection to 6G communications and autonomous systems.

Photonic integration technology offers a comprehensive solution for creating energy-efficient and mass-manufactured sensing, transmission, and computing functions in a compact chip form. Integrated photonics is well-known for its high throughput and low latency in data transmission. Generally, this technology utilises existing microelectronics infrastructure to manufacture photonic chips on wafer sizes ranging from 2 inches to 300 mm. Integrated photonics is positioned as a crucial technology that will support the microelectronics industry during the AI-driven expansion. For example, at the cloud level, integrated photonics provides energy-efficient, high-throughput, low-latency connections for inter- and intra-datacenter communication and optical circuit switching. At the edge level, integrated photonics enables data collection through photonic sensors, facilitates low-latency transmission using optical interconnects, and supports lightweight processing on edge AI devices, promoting a "process-as-it-flows" approach rather than a "store-and-process" method.

For integrated photonic, technological challenges go beyond hardware integration (compatibility). Integration of photonics in new and future CMS will require a suitable co-design strategy (including associated methods/tools) of the various technologies to be combined. A high level of integration also requires a special focus on thermal management, optimized thermal design, and suitable cooling concepts. This aspect, as well as other requirements, will determine the design and technology of future packages of photonic components. A particular focus is on those cases in which photonics and electronics have to work together at high bandwidths, such as very high-capacity transceivers, interfaces to electronic switching circuits in communications and data systems, high pixel-count active sensors, 3D imaging and displays. Another technical challenge is the lack of standard solutions for hybrid integration, as well as for development (e.g. qualification testing) or even legal framework conditions (e.g. for the regular use of smart glasses and similar applications).

Heterogeneous integration technologies are heavily driven by consumer applications, particularly portable and handheld devices. The high-volume manufacturing and supply chains for these products are predominantly located in Asia; therefore, for PFI, Europe must strengthen its own supply chain for integration and packaging solutions. In consumer electronics, the convergence of sensing and imaging—as seen in face recognition and augmented reality (AR) systems using consumer LiDAR—demands the co-integration of high-speed electronics and integrated photonics into compact, high-performance systems. Portable devices also increasingly adopt flexible structures and technologies, aiming for thinner, lighter, and more adaptable electronic components. Applications include advanced displays, wearable devices, and next-generation human–machine interfaces (HMI). Structural and 3D electronics make it possible to embed electronics into 3D surfaces and mechanical components using techniques such as moulding, additive manufacturing, or laser direct structuring (LDS). To enable these advancements, new flexible and stretchable substrates—such as thermoplastic polyurethane (TPU) and polydimethylsiloxane (PDMS)—are being developed, alongside novel active materials. These include conductive and dielectric inks based on organic compounds, metal oxides, nanomaterials, and 2D materials, supporting a wide range of applications from touch panels and RF antennas to control electronics, embedded lighting, and integrated sensors/actuators.

One of the key application drivers for the PFI of smart CMS is the IoT and its sensor nodes, which require a wide range of sensor and actuator functionalities, combined with data processing and wireless communication, and with power autonomy provided by energy storage and harvesting devices. Without an adequate alternative to primary batteries, IoT will not meet in full its deployment expectations. This also requires the development of low-power solutions for sensors and actuators, as well as radio communication components and processing. Thermal management challenges introduced by increased functionality in a minimum form factor need to be solved. New and improved energy storage, especially low-leakage rechargeable storage devices, in many cases capable of operating in harsh environments (temperature, ingress, in-vivo) needs to be developed as well as universally deployable and scalable harvesting solutions to improve the case-specific devices used today. In addition to this, it is important to improve in low-power techniques at the system level with co-design of hardware and software and overall system architecture, e.g. in wireless sensor networks, to ensure reliable, sustainable and energy efficient data collection and processing systems attending to the energy available and the possible processing capabilities split at the edge, gateway and cloud. Reliable and fault-tolerant wireless networks are required for applications where long-term continued sensing is critical (e.g. structural health monitoring of civil infrastructure).

Another domain strongly reliant on PFI are Electronic Control Units (ECUs). Due to their complexity and high degree of integration, these systems benefit from advances in generative design, miniaturization, scalability, increased processing power, cybersecurity, AI and machine learning integration, prognostics and health management as well as sensor fusion.

Components to provide power efficient computational resources, i.e. low-power microprocessors and devices with novel computational architectures such as neuromorphic devices, are needed, as are low-power computational methods, including distributed and low-power AI solutions in hardware, software, and in-sensor and data fusion data processing. In addition, reliable, energy-efficient, scalable, low-loss interconnection and packaging solutions are a necessity.

Smart CMS leverage a multitude of materials, such as silicon and other-than-silicon semiconductors, precious and rare earth metals, ceramics, polymers, glass, inks and functional materials for sensing, actuation and energy harvesting, as well as hybrid combinations of substrates and materials (e.g. Si, ceramic, polymer, glass, metallic glass), in packages and in systems, extending the coverage of the usual materials in semiconductor-based technologies. Many of the new features required by future smart systems can only be achieved by introducing novel materials into the devices and systems, from back-end of the line processing of the microchips, or by post-processing on CMOS, to novel IC carrier technologies, including fan-out wafer level packaging and other SiP technologies as well as PCBs or in printing, additive manufacturing or other means. The development of new materials and the compatibility of those (with regard to e.g. process compatibility, environmental compatibility) is critical to the future development of PFI.

Spintronics is an emerging field that holds significant potential to become a key component of future chips technologies beyond CMOS. Unlike conventional electronics that utilizes electric charges, spintronic devices leverage the spin of electrons to generate, detect and convert charge currents. The high-performance and ultra-low power consumption intrinsic to spintronics make it a revolutionary solution to the increasing demands for computing power that conventional electronics struggle to address, in particular with the increasing demand of AI. By combining sensing, processing, storage and logic on a unified platform, spintronic devices can advance technologies such as quantum computing, neuromorphic electronics and energy harvesting, significantly contributing to low-carbon prospects.

Quantum technology introduces a new modality within More-than-Moore innovations—one that was previously impractical for industrial use due to limitations such as extreme cooling requirements and poor field deployability. In recent years, rapid progress has significantly advanced the technological readiness of multiple quantum domains:

• Quantum sensors, particularly those using gas-cell technology, are undergoing industrialization and miniaturization, offering enhanced stability and accuracy in applications such as magnetometry and inertial measurement.
• Quantum computing continues to evolve, with systems becoming larger and more powerful each year.
• Quantum communications, notably quantum key distribution (QKD), enable ultra-secure encryption based on the quantum states of photons.

The expanding applicability of quantum technology stems not only from advances in materials, fabrication processes, and quantum science itself, but also from enabling technologies. These include high-precision control and low-noise readout electronics, advanced packaging, heterogeneous integration, and cryogenic cooling—many of which build directly on the achievements and processes of the electronics and semiconductor industries. As such, the continued development of these enabling technologies must be a priority.

The following key focus areas address the multimodality of ECS, which goes beyond semiconductor technology, requiring advanced packaging and heterogeneous integration of diverse materials, components and platforms. Full coverage of the physical-functional integration requires considering both the physical integration technology platforms and the functionalities of the integrated systems.

• Sensing, imaging and actuation:
• Sensors and actuators leveraging the integration of MEMS/NEMS, MOEMS and micro-optics elements.
• Sensors and actuators for biological, medical and diagnostic applications, and for sensing of human vital signs and biomarkers, as well as for selective detection of gas and volatiles, allergens, residues/pollution in food/water, atmospheric particles, hazardous substances and radiation.
• Innovative sensor technologies to improve the performance of analytical equipment used in materials science, including battery and hydrogen energy ecosystems, and the semiconductor industry resulting in more effective research, development and manufacturing, including yield and quality control.
• Sensors and systems enabling integration behind OLED and sensing through the OLED display.
• Smart surfaces for HMI including integrated sensors for micro-LED and mini-LED displays.
• Sensors and systems for sensing and imaging in the short wavelength infrared range, based on Ge on Si Integration or Pb-free quantum dots, enabling a broad range of new applications in the areas of lighting, biotechnology and life sciences, photovoltaics and information processing.
• Silicon-integrated mmWave and THz sensors for high-resolution imaging, spectroscopy, and near-field radar, enabling scalable, low-cost, light-weight, and miniaturized smart sensing systems for industrial, biomedical, security, automotive and space applications
• Advanced global shutter and rolling shutter CMOS Image Sensors (CIS) based on novel pixel technology for VIS-NIR imaging applications, with the goal to improve the performance, sensitivity, and efficiency of VIS-NIR CIS pixels, enabling high-quality imaging across a broad spectral range, as well as infrared imaging.
• Advanced spintronic sensors, particularly TMR-based devices, for ultra-high sensitivity, low-power magnetic field sensing with seamless CMOS and edge AI accelerator integration to enable high-performance, ultra-low-power solutions for computing, memory, and sensing.
• Imaging systems: lidars, radars and sonars, X-ray and magnetic resonance, including multi-modal and hyperspectral, i.e. spectrally resolved, sensors.
• Sensors and systems utilizing quantum principles, e.g. single photon sensors, including required cryogenic and cooling components and systems.
• Devices with new features and improved performance for sensing and actuation using novel materials (metal nanowires, carbon nanotubes (CNTs), graphene and other 2D materials, spintronics materials, cellulose nanofibers, nitrogen vacancies in diamond, metamaterials, metallic glass etc.), in combination with, or integrated on, CMOS.
• Sensor fusion and virtual sensors including appropriate data and communication infrastructure, e.g. for condition monitoring, prognostics and health management for ECUs.
• Ultra-low power event-based sensors, e.g. inertial motion & image detection for asset tracking, incident/anomaly detection, etc.
• Dedicated LED and laser light sources from UV to visible to IR for sensing applications by avoidance/replacement of hazardous material (e.g. Cd/Pb-free QDs for color conversion, UV-C LEDs instead of Hg-lamps).
• Materials that ensure the hermetic sealing of sensors, actuators and systems and at the same time contribute to the miniaturization of components.
• MEMS technology
• CMOS or GaN-compatible thin film piezoelectric materials, such as ScAlN for piezo-actuated MEMS sensors and actuators.
• Acoustic piezo-MEMS devices, pMUTs, and acoustic RF filters for high frequencies above 6 GHz.
• Audio MEMS systems such as micro speakers with more advanced integrated functionality, such as amplifier and integrated noise canceling (ANC).
• MEMS for medical devices and consumer health, including Point-of-care diagnostic systems with reduced complexity and faster time-to-results.
• Miniaturized sensors for integration into everyday devices for continuous health monitoring and overall health evaluation.
• Integrated Photonics
• Handling of materials like Lithium Niobate and Indium Phosphide that allow for the integration for active photonic devices in electronic components for improved performance, such as higher bandwidth in modulators, and detectors.
• New waveguide materials and components to expand the wavelength range from UV up to mid IR optical elements for beam shaping and manipulation (like ultra-thin curved waveguides, meta-lenses, tunable lenses and filters, next generation holograms, ultra-wide-angle holograms).
• High-speed modulators play a crucial role in high-speed optical transceivers, which are vital for optical connectivity in networking and computing. On the other hand, low-speed modulators are used in various applications, including datacenter transceivers, photonic quantum computing systems, programmable photonic meshes, photonic neuromorphic computing systems, optical circuit switching systems, and optical beamforming systems, among others.
• Microbolometer thermal imagers with integrated optics and edge AI recognition for safety and mobility applications such as Automated Emergency Braking (AEB) and sensing for autonomous vehicles. Pixelated light sources for illumination, visualization and communication purposes including vertical hetero-integration of III-V materials on Silicon.
• Display technologies (like OLED-on-silicon, micro-LEDs, MEMS-mirrors, Phase parrays) and sensors (e.g. for eye tracking). Currently, commercial head-mounted display solutions rely upon high-end microdisplays deploying various technologies such as organic light-emitting diodes (OLEDs), micro-LEDs, lasers, and LEDs with liquid crystal on silicon (LCoS), or microelectromechanical systems (MEMS). Nonetheless, these microdisplays will still need to undergo further improvements regarding size, power consumption (<300 mW; full display system), device efficiency, resolution (8K and beyond, pixel densities > 10kppi), high dynamic range (HDR), colour gamut, contrast and refresh rate. Besides utilizing classical visualization modes, plenoptic modes such as light field, multifocal and holography are gaining ground, holding the promise for truly immersive experiences and potentially improving vergence-accommodation behavior for (Head Mounted Device) HMDs and autostereoscopic displays. Both the classical approach and the plenoptic solutions require not only a significant evolution regarding the capabilities and features of the opto-nanoelectronic display devices – such as multi-ridge lasers and associated waveguide combiners, light-field (micro)displays, multifocal light engines, spatial light modulators (SLMs) based on LCOS, and MEMS – but also state-of-the-art drivers and optical systems and architectures.
• New devices for Quantum PICs: New devices, materials and heterogenous integration concepts for photonics-based quantum computation, communications and sensing applications. For example, integrated and compact single photon emission, manipulation and detection at room temperatures. Integration of broadband or programmable wavelength light sources with silicon photonics platforms.
• Flexible electronics
• Sensing devices and power sources compliant with hybrid integration in wearables, considering flexibility, durability (e.g. washability) and biocompatibility.
• Flexible and stretchable sensors and modules, e.g. OLED displays, OPVs, touch surfaces and other sensors/actuators, conformal antennas.
• Functional materials for flexible and stretchable devices; organic and inorganic semiconductor materials and inks, perovskites for OPV.
• Barrier materials, dielectrics, and transparent conductor materials and inks for flexible electronics and additive manufacturing.
• New materials and manufacturing processes for 3D and structural electronics, such as laser direct structuring (LDS) and moulding, with novel substrates like TPU and PDMS, conductive/dielectric inks, and spintronics materials.
• Hybrid integration of discrete components or even bare chips on flexible/stretchable wiring substrates for applications such as wearable electronics
• Communications
• Module-level high-speed wireless communication features, including current and new frequency bands.
• Smart and secured communications based on spintronics and magnonics nano-oscillators
• High-speed photonics communications modules beyond 1 Tb/s.
• New front-end components, filters and functionalities, e.g. active antennas for 5G and 6G communications and non-terrestrial network solutions.
• Low latency and low power communications in-package/module as well as at system level for the edge and IoT devices.
• Energy and application context aware comms that can dynamically adapt the architecture (edge, gateway, cloud) to meet sensing, processing and actuation needs based on limited energy available, especially at the edge. Continuous delivery of new features and fixes through Over-The-Air (OTA) updates to ensure the security of the device over time and reduce the digital waste by increasing the life span of a device.
• Strategies and components for Electromagnetic interference (EMI) mitigation and reliable operation in harsh environmental conditions.
• Energy and thermal management
• Low-power/low-loss modules for low-power sensing, actuation, processing and communication.
• Multi-source power architectures with digital interfaces driven by dynamic and context aware algorithms that can adapt based on energy available versus needed for sensing, actuation, processing and communication.
• Energy-autonomous multi-sensor modules and systems including energy harvesting, sensing, actuation, processing and communication.
• Power management components and modules compatible with harsh environments (high temperatures, vibrations, bio-compatibility, ingress, electromagnetic interference (EMI) conditions for industrial, automotive, med-tech and space technology).
• Devices using non-toxic materials for efficient energy sources, storage and harvesting devices (thermoelectric, piezoelectric, tribo-electricity, etc.), and higher performing electrodes and electrolytes for improved capacity and low leakage of energy storage devices or new lightweight energy harvesters for mobility and transportation applications.
• Solutions for thermal management for integrated photonics and RF systems at different integration levels including advanced and active cooling systems.
• Thermal management and smart cooling systems for industrial applications and harsh environments.
• Efficient smart compact cooling solutions and approaches for quantum devices and cryogenic multiplexing with semiconductors or superconducting devices.
• Information processing
• Component and system-level features for self-diagnosis and module-level signal processing and control features for self-diagnosis, self-monitoring, and self-learning and self-repair.
• Sensor level hardware and software solutions for security and privacy and data reliability.
• Machine learning and artificial intelligence and data analysis at the sensor, module and systems level, i.e. on the edge data analysis embedded at different levels for smarter devices, including AI at sensor level.
• Integrated and scalable solutions, both Software and Hardware integration with increased processing power for more sophisticated features, especially for edge AI and ECUs.
• Use of low power computing based on in-memory computing and bio-inspired spintronics devices for data analysis Hardware-software co-design to enable secure, low-level functionalities such as communication, visualization, and security protocols.
• Integration of electronic components in electronic control units:
• Another domain strongly reliant on PFI are Electronic Control Units (ECUs). Due to their complexity and high degree of integration, these systems benefit from advances in generative design, miniaturization, scalability, increased processing power, cybersecurity, AI and machine learning integration, prognostics and health management as well as sensor fusion.
• Sensing, imaging and actuation

Sensor fusion and virtual sensors including appropriate data and communication infrastructure, e.g. for condition monitoring, prognostics and health management for ECUs.

• Information processing

Integrated and scalable solutions, both Software and Hardware integration with increased processing power for more sophisticated features, especially for edge AI and ECUs

• Advanced packaging and SiP technologies

Robust heterogeneous 3D integration of sensors, actuators, electronics, processing units, communication, RF front-end components and energy supply into miniaturized systems.