JTJens Trommer

Jens Trommer

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Jens Trommer has a total of 198 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Enabling High‐Frequency Characterization for Dynamic Modeling of Schottky Barrier Field‐Effect Transistors

    by Bruno Neckel Wesling, Cigdem Cakirlar, Giulio Galderisi, Chhandak Mukherjee, Magali De Matos, Cristell Maneux, Thomas Mikolajick, Jens Trommer, Marina Deng

    This work presents a comprehensive methodology describing static and dynamic characterization and compact modeling of an ambipolar Schottky barrier field‐effect transistor (SBFET) technology, taking into consideration the integration of specifically designed test structures necessary for accurate scattering parameter measurements. In this work, we demonstrate, for the first time, the complete design flow for this emerging technology, starting with design and fabrication of dedicated high‐frequency test structures for scattering‐parameter measurement up to 110 GHz. The high frequency data are then used to validate a physics‐based compact model for the ambipolar SBFETs. Physically meaningful results are obtained with a good agreement between the measured and compact model data, demonstrating the usefulness of such dedicated methodologies for the refinement of the emerging technology performance.

  • Electrostatics of Common‐Channel Reconfigurable Field‐Effect Transistors: From X‐Shape to U‐Shape

    by Cigdem Cakirlar, Tetiana Obukhova, Oskar Baumgartner, Mischa Thesberg, Jose Maria Gonzalez‐Medina, Sara Mannaa, Ian O’COnnor, Thomas Mikolajick, Damien Deleruyelle, Jens Trommer

    This work investigates the effect of scaling on the electrostatic behavior of cross‐shape reconfigurable field‐effect transistors (RFETs) using three‐dimensional technology computer‐aided design simulations. The emerging device concept consists of multiple nanowires in a junction of a common channel, where multiple gates are aligned on the silicon and silicide junction interface, enabling polarity switching and signal routing within a compact layout. Although such a geometry can be utilized for new directions in neuromorphic or analog computation, it significantly increases the electrostatic interactions. The focus of the work is to understand the impact of the gate‐to‐gate distance and nanowire width on the on‐state current. The energy barrier in the ungated common‐channel region was found to be the limiting factor, as excessive fringing field coupling at short gate‐to‐gate distances causes electrostatic crowding and barrier sharpening, thereby restricting carrier transport. To overcome the electrostatic limitation of the cross‐shape RFET, a three‐dimensional U‐shape RFET was proposed. Due to its gate‐all‐around coverage of the metal–semiconductor junctions, electrostatic control was found to be stronger. U‐shape geometry achieves a high on‐state current with a minimal sensitivity to gate‐to‐gate distance variations. This work derives practical design guidelines for 3D RFET architectures, aiming to balance device compactness and electrostatic efficiency.

  • Open Access

    Re‐Thinking Electronic Circuit Design Through Emerging Reconfigurable Field Effect Transistors

    by Jens Trommer, Juan P. Martinez, Yuxuan He, Giulio Galderisi, Katrin Pingen, Eduardo Sergio Oliveros‐Mata, Lisa Tellis, Cigdem Cakirlar, Christoph Beyer, Brojogopal Sapui, et al.

    Reconfigurable field‐effect transistors, short RFETs, are a unique class of emerging electronic devices that merge n ‐type, p ‐type, and ambipolar operation modes into a single element. Unlike conventional MOSFETs, the dominant carrier type in RFETs is purely defined by electrostatics and can be dynamically changed at runtime. This enables new design possibilities by enabling post‐fabrication functionality definition, thereby increasing circuit density and adaptability without requiring additional feature‐size scaling. However, the added value of the devices requires a reassessment of classical design paradigms to unlock their full potential. This paper aims to provide a comprehensive review of the state of the art in RFET circuit design. Starting from the underlying device physics and material systems, such as group‐IV and two‐dimensional (2D) materials, the manifold circuit design possibilities in digital, analog, and post‐Shannon computing are introduced. We highlight how RFETs aim to enable simplified, energy‐efficient, and reusable hardware for edge computing, hardware security, bio‐sensing, and cryogenic electronics. Thereby, device‐ and circuit‐level results are exemplified by measurements of transistors fabricated in an industrial 22 nm FDSOI technology. The paper also provides an overview of future key challenges in reliability, BEOL integration, and electronic design automation, offering an outlook on the path towards scalable, sustainable reconfigurable electronics.

  • Open Access

    Gigahertz Cutoff Frequencies and High Gain in Graphene-Based Hot-Electron Transistor Enabled by Material Engineering

    by Carsten Strobel, André Heinzig, Andre Hiess, Martin Knaut, Md Tarik Hossain, Andrey Turchanin, Tilo Meister, Frank Ellinger, Jens Trommer, Viktor Havel, et al.

    Graphene (Gr)-based hot-electron transistors (GHETs) offer high potential for high-frequency applications due to the extremely thin nature of graphene as a base material. For the first time we present GHETs with excellent DC device characteristics and gigahertz operation achieved by an optimization of the emitter-base (E/B) composition. The optimized E/B composition enables more efficient injection of hot electrons into the base, thereby improving charge transport and reducing scattering losses. As a result, a record-high measured common-emitter current gain beta of 42 was achieved using a SiO2/Gr E/B structure. Moreover, when using a MoS2/Gr E/B junction the maximum output current is increased to record values of approximately 2000 A/cm2, representing a significant improvement in performance over previous devices. Furthermore, cutoff frequencies of close to 1 GHz are determined for nonoptimized SiO2/Gr-based devices. The experimentally observed device characteristics are very promising for future high-speed nanoelectronics.

  • Open Access

    Efficient In‐Hardware Matrix–Vector Multiplication and Addition Exploiting Bilinearity of Schottky Barrier Transistors Processed on Industrial FDSOI

    by Juan P. Martinez, Niladri Bhattacharjee, Yuxuan He, Giulio Galderisi, Brojogopal Sapui, Sara Mannaa, Ian O'COnnor, Violetta Sessi, Peter Javorka, Thomas Mikolajick, et al.

    Machine learning and Artificial Intelligence (AI) tasks have stretched traditional hardware to its limits. In‐hardware computation is a novel approach that aims to run complex operations, such as matrix–vector multiplication, directly at the device level for increased efficiency. This work shows that the current of a fully CMOS‐compatible Schottky barrier transistor can respond linearly to changes in either the source–gate or the source–drain voltage. This bi‐linearity allows analog vector multiplication directly at the device level. Alongside the access to independent biasing through the additional back‐gate offered by the FDSOI technology, which can be used to implement addition directly at the device level, it has great potential for in‐hardware computation. The performance and multilevel operation are demonstrated by the electrical characterization of a two‐transistor system, showcasing how the technology could be implemented in larger crossbar arrays. The potential for in‐hardware computation is evaluated against a full in‐software solution by comparing the inference performance on the Iris dataset. The simulated in‐hardware schemes utilizing the back‐biased Schottky barrier transistors can achieve the same accuracy as the digital implementation but with an order of magnitude less power dissipation per operation when compared to CMOS‐based digital hardware accelerators proposed in the literature.


  • Open Access

    A Compendium of Logic Gates Based on Reconfigurable Three‐Independent‐Gate Transistors Realized in FDSOI Hardware

    by Juan P. Martinez, Yuxuan He, Giulio Galderisi, Violetta Sessi, Niladri Bhattacharjee, Peter Baars, Kerstin Poenisch, Annekathrin Zeun, Konstantin Li, Fernando Koch, et al.

    This work presents the electrical characterization of sixteen different logic gates built entirely from three‐independent‐gate reconfigurable transistors. The circuits are fabricated on full‐scale 300 mm wafers using the industrial 22 nm fully depleted silicon‐on‐insulator process of GlobalFoundries, with only minimal modifications to the baseline CMOS flow. The demonstrations include a reconfigurable 2‐2 AND‐OR‐Inverter gate and a fully functional 1‐bit adder comprising eight transistors. Quasi‐static and transient on‐wafer measurements confirm correct functionality and provide insight into the frequency limitations imposed by the current design and test setup. Finally, to explore scalability, a ripple‐carry adder is simulated based on the experimentally realized 1‐bit adder, illustrating how scaled devices and optimized layouts could enable low‐power, CMOS‐compatible applications.


  • Open Access

    Junctionless Silicon Nanowire Transistors without the Use of Impurity Doping

    by Soundarya Nagarajan, Dirk König, Ingmar Ratschinski, Giulio Galderisi, Somayeh Shams, Thomas Mikolajick, Daniel Hiller, Jens Trommer

    With the shrinking dimensions of semiconductor structures reaching the nanoscale, conventional impurity doping techniques face several challenges due to their statistical nature, temperature dependence, and degradation in efficiency of the doping method. In addition, the cryogenic operation of highly doped transistors is complicated due to carrier freeze-out, which significantly reduces the availability of mobile charges, degrading device performance and inducing noise. Here, an innovative material solution is presented that enables silicon nanowire junctionless transistors without requiring impurity doping within the active semiconductor region. To this end, a SiO2 dielectric shell with deliberate defect engineering surrounding both the channel and the contact regions - known as direct modulation doping─is used to modify the nanoscale transport properties of the silicon. The obtained active carrier densities in the experiment are comparable to highly impurity-doped devices in the range of ∼1018⁡cm-3 and remain stable over a broad temperature range from 400 K down to 77 K. The primary advantage of removing dopants from the channel is evident in the enhanced field-effect mobilities, which increase from 115 to 331 cm2V-1s-1 as temperature decreases. The fabricated nanowire transistors in this work provide a high on/off ratio of ≥106, and a stable on-state performance down to 77 K. Hybrid-density-functional-theory calculations are carried out to show that there are no fundamental roadblocks to employing the method to devices with ultrascaled dimensions. The device architecture is positioned for applications in energy-efficient cryo-electronics and quantum technologies by addressing the limitations associated with conventional impurity doping.


  • Open Access

    Ultra-Low Voltage Reconfigurable FETs in 22nm FDSOI Technology Enabling Dynamic Circuit Obfuscation for Embedded Security

    by Yuxuan He, Aniruddh Holemadlu, Niladri Bhattacharjee, Giulio Galderisi, Violetta Sessi, Juan Martinez, Mark Wijvliet, Shubham Rai, Viktor Havel, Peter Baars, et al.

    Reconfigurable Field-Effect Transistors are fully CMOS-compatible emerging devices capable of changing between n-type and p-type characteristics during runtime. This inherent polymorphism extends to the logic gate level, enabling circuit obfuscation beyond what can be achieved with traditional CMOS devices. In order to exploit these capabilities, especially for embedded security solutions at the edge, it is crucial that the devices operate at low voltages to minimize power consumption. This work presents the first demonstration of the highly expressive three-independent gate reconfigurable transistor variant operated at 0.8 V. The devices are processed based on the 22nm FDSOI technology of GlobalFoundries and are fully co-integratable with standard CMOS devices. We leverage the self-dual nature of these devices to implement a circuit obfuscation scheme to merge two individual circuits into a single one, thus illustrating a scalable method that can be extended to larger cryptographic engines to effectively mask their functionality from an outside attacker. The functionality of the method is validated by simulating a merged adder/subtractor circuit in Cadence Virtuoso, utilizing a previously developed Verilog-A model of the scaled devices. This work demonstrates the feasibility of RFETs operating at supply voltages as low as 0.8 V with symmetrical P and N on-states, enabling energy-efficient hardware security applications, such as reconfigurable adder/subtractor circuits and polymorphic logic blocks.


  • Design and Evaluation of an RFET Standard Cell Library Compatible with 22 nm FDSOI

    by Nima Kavand, Armin Darjani, Niladri Bhattacharjee, Jorge Navarro Quijada, Juan P. Martinez, Thomas Mikolajick, Jens Trommer, Akash Kumar

    Reconfigurable Field-Effect Transistor (RFET) is an emerging technology that offers runtime reconfigurability between n-type and p-type operation modes. Its unique features, including intrinsic polymorphism, symmetric I–V characteristics, multiple independent gate control, and low leakage current, unlock new possibilities for designing energy-efficient logic circuits, particularly polymorphic ones well-suited for hardware security techniques such as logic locking. Although various RFET device realizations with different materials and structures have been proposed, only a few studies have developed standard cell libraries that enable the evaluation of large-scale circuits implemented with these technologies. Moreover, the existing libraries are not designed according to the design rules of well-established industrial processes. This paper presents standard cell libraries comprising both static and reconfigurable logic cells, enabling a complete RTL-to-GDSII flow for Three-Independent-Gate RFETs (TIG-RFETs), compatible with an industrial 22 nm FDSOI technology. We implement the circuits from the ISCAS’85 and EPFL’15 benchmark suites using our libraries within a commercial EDA tool flow and compare the synthesis results with their CMOS counterparts in the same technology node in terms of Power, Performance, and Area (PPA). The results demonstrate that TIG-RFETs can significantly reduce power consumption, though with a performance trade-off that limits the achievable clock frequency to a few MHz. Furthermore, we analyze the results to highlight the advantages and limitations of this implementation of TIG-RFETs and discuss potential future improvements.


  • Modulation Acceptor-Doping as an Alternative to Classical P-Type Impurity Doping for Silicon

    by Soundarya Nagarajan, Ingmar Ratschinski, Stefan Schmult, Thomas Mikolajick, Daniel Hiller, Jens Trommer

    The statistical nature of impurity dopants introduced in a semiconductor leads to variability in electrical properties, posing significant challenges to device performance. This randomness becomes especially problematic as device dimensions shrink. Alternative approaches for material modification that offer precise control over the material properties, including predictable behavior at low temperatures, are essential for largescale device integration. Modulation acceptor doping in SiO2 was introduced as a nanoscale-doping alternative to classical impurity doping for silicon, where the carriers are spatially isolated from the dopants, resulting in a drastically reduced Coulomb scattering within silicon. In this work, the stability of the new doping method in ambient conditions is assessed. Further, a comparative analysis of impurity (p-type) doped and modulation acceptor doped Hall structures is presented. The modulation doped sample series reveals a consistent resistivity and a stable carrier density. While, the impurity-doped sample series demonstrates a lower dopant activation efficiency, higher resistivity and highlights the statistical nature of the classical method. In addition, the temperature behaviour of resistivity is studied from 300 K down to 100 K.