arXiv subjects

Mahmoud Rasras

Publications and source records attributed to Mahmoud Rasras.

15 recordsLinked to original sources

arXiv2026

MDTransformer: A Hardware-Software Co-Design of Mode-Division Photonic Transformer Accelerator with Inverse-Designed Coherent Crossbar

Recently, photonic transformer accelerators (PTAs) have successfully achieved significant speedup and energy efficiency improvements over electronic accelerators for expediting Transformer inference. However, state-of-the-art rely on expensive multi-wavelength light generation and large dot-product units due to active phase-shifter components, thus making their approach inefficient and impractical. To address this, we propose MDTransformer, a novel hardware-software co-design of PTA based on mode-division optical dataflow and operations. Specifically, MDTransformer performs complex matrix operations using spatial-mode interference, that leverages the inverse-designed multi-mode couplers, crossings, and Mach-Zehnder IQ modulators into a compact mode-division photonic tensor core (MPTC), capable of executing matrix multiplications in the optical domain. Its each guided mode (i.e., TE0-TE3) acts as an independent computational lane, enabling four-fold parallelism-per-waveguide without spectral filtering or free-spectral-range limitations. Moreover, its coherent detection and IQ modulation jointly encode amplitude and phase, realizing complex-valued arithmetic for full-range operations in transformers. MDTransformer offers analog multiplication with sub-4-bit effective precision and inter-modal crosstalk below -30 dB. Its inverse-designed approach also offers scalable and full compatibility with single-laser continuous-wave operation at 1550 nm. Experimental results show that MDTransformer achieves 40.4% area reduction, 63.6% power saving, 40.6% energy saving, and comparable latency over the state-of-the-art PTA across different workloads (i.e., DeiT-Tiny/Small/Base and BERT-Base/Large). These results show that MDTransformer offers a practical solution for high-performance and energy-efficient transformer-based systems.

cs.AR↗

arXiv2026

Programmable photonics enabled by ferroionic two-dimensional materials

Artificial intelligence (AI) models are scaling rapidly, exposing fundamental limitations in conventional computing architectures, where the physical separation of memory and computation imposes substantial energy and latency overheads. Optical neural networks offer a viable solution by enabling high-throughput, low-latency computation through the intrinsic parallelism of light. However, the scalability of photonic computing is constrained by the lack of materials that can provide low-loss, energy-efficient, and nonvolatile control of optical phase. Here, we discuss the emerging role of ferroionic two-dimensional materials as a platform for programmable photonics. Unlike conventional ferroelectrics, where polarization arises from bounded lattice distortions, ferroionic systems enable field-driven ionic redistribution, providing access to a continuum of stable and reconfigurable states. We outline how these material dynamics can be exploited to realize multi-level photonic states and nonvolatile phase control. Furthermore, we highlight a back-end integration strategy that introduces multifunctionality, including tunable optical nonlinearities into pre-fabricated photonic circuits without compromising optical performance. By bridging material-level dynamics with device- and system-level functionality, ferroionic materials provide a pathway toward reconfigurable, low-loss, and energy-efficient photonic computing architectures, opening new opportunities for neuromorphic and adaptive photonic systems.

physics.optics↗

arXiv2026

DxPTA: An Architecture Design Space Exploration with Optical Dataflow-guided Strategy for HW/SW Co-Design of Photonic Transformer Accelerators

Transformer-based networks have emerged as prominent AI models with state-of-the-art performance, which potentially pave the way toward artificial general intelligence (AGI). However, their large sizes still hinder their efficient implementation, thus highlighting the need for alternate solutions to enable their energy-efficient acceleration. Recently, state-of-the-art works propose photonic transformer accelerators (PTAs) with significant speedup and energy efficiency improvements over the conventional electronic accelerators. However, their PTA architectures are developed without considering the application constraints (e.g., area, power, energy, and latency). Moreover, their manual design approach also requires huge design time to determine a suitable architecture for the targeted application, hence making this approach not scalable. To address these limitations, we propose DxPTA, a novel design space exploration methodology for enabling efficient hardware/software co-design of the appropriate PTA architecture that meets all constraints. It is achieved by (1) identifying the PTA architecture parameters based on the coherent optical dataflow; (2) analyzing the impact/significance of the parameters; and (3) leveraging this analysis for devising a constraint-aware architecture search algorithm. Experimental results show that, our DxPTA can find the appropriate PTA architectures for different transformer-based models (i.e., DeiT-T/S/B and BERT-B/L). It achieves up to 26mm^2 area, 4.8W power, 39mJ energy, and 6ms latency, for constraints of 50mm^2 area, 5W power, 50mJ energy, and 10ms latency; with 15.2x faster searching time than the exhaustive approach. These results demonstrate the potential of DxPTA methodology for enabling efficient PTA designs for diverse AGI-based applications.

cs.AR↗

arXiv2025

Polarization Control and TM-Pass Filtering in SiN Photonics Integrated with 2D Multiferroic Materials

Polarization control is a critical function in integrated photonic circuits, directly impacting performance, stability, and signal integrity. In this work, we demonstrate the integration of multiferroic two-dimensional (2D) material CuCrP2S6 (CCPS) with silicon nitride (SiN) photonic devices to achieve polarization-selective filtering and rotation. Our experimental results show that microring resonators incorporating CCPS exhibit transverse magnetic (TM)-pass filtering with a polarization extinction ratio exceeding 25 dB and a low insertion loss of ~ 0.2-0.4 dB at 1500-1600 nm. Additionally, under TM-mode input, straight waveguides loaded with CCPS can achieve a significant polarization rotation, with azimuth angle shifts reaching up to 92.9{\deg}. Simulations and experimental validation indicate that the primary mechanism behind these effects is the polarization-dependent optical mode overlap, governed by the refractive index profile of the CCPS/SiN hybrid system and the waveguides' geometrical dimensions. The anisotropic properties of CCPS provide further enhancement but play a secondary role. These results highlight the potential of CCPS-integrated devices for compact, high-performance polarization control in on-chip photonic platforms.

physics.optics↗

arXiv2024

Non-Reciprocal Response in Silicon Photonic Resonators Integrated with 2D CuCrP2S6 at Short-Wave Infrared

Achieving non-reciprocal optical behavior in integrated photonics with high efficiency has long been a challenge. Here, we demonstrate a non-reciprocal magneto-optic response by integrating multilayer 2D CuCrP2S6 (CCPS) onto silicon micro ring resonators (MRRs).Under an applied magnetic field, the CCPS intralayer ferromagnetic ordering, characterized by easy-plane magneto-crystalline anisotropy, induces asymmetrical modal responses in the clockwise (CW)and counterclockwise (CCW) light propagation directions. The proposed configuration achieves a low insertion loss of 0.15 dB and a high isolation ratio of 28 dB at 1550 nm. Notably, it exhibits a significant resonance wavelength splitting of 0.4 nm between the counter propagation directions, supporting a 50 GHz optical bandwidth. Operating directly in the transverse electric (TE) mode, it aligns with the main polarization used in silicon photonics circuit, eliminating the need for additional polarization management. The device is ultra-compact, with 2D flake interaction length ranging from 22 um to 55 um and a thickness between 39 nm and 62 nm. Its operation range covers the entire C-band with a bandwidth of up to 100 nm. These attributes make our hybrid CCPS/Si device ideal for advanced non-reciprocal optical applications in the short-wave infrared (SWIR)spectrum, crucial for enhancing the resilience of optical systems against back-reflections.

physics.optics↗

arXiv2023

Unlocking Electro-Optic Tuning in Hybrid Silicon Photonics Based on Ferroionic 2D Materials

Tunable optical materials are indispensable elements in modern optoelectronics, especially in integrated photonics circuits where precise control over the effective refractive index is essential for diverse applications. Two-dimensional materials like Transition Metal Dichalcogenides (TMDs) and graphene exhibit remarkable optical responses to external stimuli. However, achieving distinctive modulation across a broad spectrum while enabling precise phase control at low signal loss within a compact footprint remains an ongoing challenge. In this work, we unveil the robust electro-refractive response of multilayer ferroionic two-dimensional CuCrP2S6 (CCPS) in the near-infrared wavelength range. By integrating CuCrP2S6 into SiPh microring resonators (MRR), we enhance light-matter interaction and measurement sensitivity to minute phase and absorption variations. Results show that electrically driven Cu ions can tune the effective refractive index on the order of 2.8 x10-3 RIU (refractive index unit) while preserving extinction ratios and resonance linewidth. Notably, these devices exhibit low optical losses of 2.7 dB/cm and excellent modulation efficiency of 0.25 V.cm with a consistent blue shift in the resonance wavelengths among all devices. These results outperform earlier findings on phase shifter based on TMDs. Consequently, CCPS integration endows characteristics akin to those of high-index active dielectric materials. Moreover, we showcase the electro-optic tuning sensitivity to light polarization, opening avenues for versatile light manipulation. The dual optoelectronic and ionotronic capabilities of the two-terminal CCPS devices hold vast potential, spanning applications such as phased arrays, optical switching, and neuromorphic systems in light-sensitive artificial synapses.

physics.optics↗

arXiv2023

Response to "On the giant deformation and ferroelectricity of guanidinium nitrate" by Marek Szafrański and Andrzej Katrusiak

Following a well-established practice of publishing commentaries to articles of other authors who work on materials that were earlier studied by them (n.b. six published comments[1-6]), Marek Szafrański(MS) and Andrzej Katrusiak (AK) have filed on the preprint server arXiv a manuscript entitled "On the giant deformation and ferroelectricity of guanidinium nitrate"[7] with comments on our article "Exceptionally high work density of a ferroelectric dynamic organic crystal around room temperature" published in Nature Communications (2022, 13, 2823).[8] Both in the submitted comment as well as in the required (by the journal) direct communication with us preceding its posting, MS and AK have expressed dissatisfaction with the choice of literature references in our article, for which they felt that their previous work on this material has not been cited to a sufficient extent. In their comment, they summarize their other remarks on our article as "the structural determinations of GN [guanidinium nitrate] crystals, their phase transitions and associated giant deformation, as well as its detailed structural mechanism, the molecular dynamics and dielectric properties were reported before, while the semiconductivity, ferroelectricity, and fatigue resistance of the GN [guanidinium nitrate] crystals cannot be confirmed."[7] Apart from the sentiments of MS and AK on our choice of cited literature, we find their comments on the scientific content of our article to be strongly biased towards their own results and unfounded. Below, we provide a detailed response to their comments.

cond-mat.mtrl-sci↗

arXiv2023

Perception, performance, and detectability of conversational artificial intelligence across 32 university courses

The emergence of large language models has led to the development of powerful tools such as ChatGPT that can produce text indistinguishable from human-generated work. With the increasing accessibility of such technology, students across the globe may utilize it to help with their school work -- a possibility that has sparked discussions on the integrity of student evaluations in the age of artificial intelligence (AI). To date, it is unclear how such tools perform compared to students on university-level courses. Further, students' perspectives regarding the use of such tools, and educators' perspectives on treating their use as plagiarism, remain unknown. Here, we compare the performance of ChatGPT against students on 32 university-level courses. We also assess the degree to which its use can be detected by two classifiers designed specifically for this purpose. Additionally, we conduct a survey across five countries, as well as a more in-depth survey at the authors' institution, to discern students' and educators' perceptions of ChatGPT's use. We find that ChatGPT's performance is comparable, if not superior, to that of students in many courses. Moreover, current AI-text classifiers cannot reliably detect ChatGPT's use in school work, due to their propensity to classify human-written answers as AI-generated, as well as the ease with which AI-generated text can be edited to evade detection. Finally, we find an emerging consensus among students to use the tool, and among educators to treat this as plagiarism. Our findings offer insights that could guide policy discussions addressing the integration of AI into educational frameworks.

cs.CY↗

arXiv2023

A High-Speed Waveguide Integrated InSe Photodetector on SiN Photonics for NIR Applications

On-chip integration of two-dimensional (2D) materials offers great potential for the realization of novel optoelectronic devices in different photonic platforms. In particular, indium selenide (InSe) is a very promising 2D material due to its ultra-high carrier mobility and outstanding photo-responsivity. Here, we report a high-speed photodetector based on a multilayer 90 nm thick InSe integrated on a silicon nitride (SiN) waveguide. The device exhibits a low dark current of 10 nA at 1V bias, a remarkable photoresponsivity of 0.38 AW-1, and high external quantum efficiency of 48.4% measured at 5 V bias. This performance is tested at near-infrared (NIR) 976 nm wavelength under ambient conditions. Furthermore, using numerical and experimental investigations, the estimated absorption coefficient per unit length is 0.11dB/um. To determine the dynamic response of the photodetector, its small and large signal frequency response are also evaluated. A 3-dB radiofrequency (RF) bandwidth of 85 MHz is measured with an open-eye diagram observed at 1 Gbit/s data transmission. Given these outstanding optoelectronic merits, active photonic devices based on integrated multilayer InSe can be realized for a variety of applications including short-reach optical interconnects, LiDAR imaging, and biosensing.

physics.optics↗

arXiv2020

Plasmon-enhanced graphene photodetector with CMOS-compatible titanium nitride

Graphene has emerged as an ultrafast optoelectronic material for on-chip photodetector applications. The 2D nature of graphene enables its facile integration with complementary metal-oxide semiconductor (CMOS) microelectronics and silicon photonics, yet graphene absorbs only $\sim$2.3% of light. Plasmonic metals can enhance the responsivity of graphene photodetectors, but may result in CMOS-incompatible devices, depending on the choice of metal. Here, we propose a plasmon-enhanced photothermoelectric graphene detector using CMOS-compatible titanium nitride (TiN) on the silicon-on-insulator (SOI) platform. The device performance is quantified by its responsivity, operation speed, and noise equivalent power. Its bandwidth exceeds 100$\,$GHz, and it exhibits a nearly flat photoresponse across the telecom C-band. The photodetector responsivity is as high as 1.4$\,$A/W (1.1$\,$A/W external) at an ultra-compact length of 3.5$\,μ$m, which is the most compact footprint reported for a graphene-based waveguide photodetector. Furthermore, it operates at zero-bias, consumes zero energy, and has an ultra-low intrinsic noise equivalent power (NEP$\,$<$\,25\:\text{pW/}\sqrt{\text{Hz}}$)

physics.app-ph↗

arXiv2020

Planar Multilayered 2D GeAs Schottky Photodiode for High Performance VIS-NIR Photodetection

Novel group IVV 2D semiconductors (e.g., GeAs and SiAs) has arisen as an attractive candidate for broad-band photodetection and optoelectronic applications. This 2D family has wide tunable bandgap, excellent thermodynamic stability, and strong in-plane anisotropy. However, their photonic and optoelectronic properties have not been extensively explored so far. In this work we demonstrate a broadband back-to-back metal-semiconductor-metal (MSM) Schottky photodiode with asymmetric contact geometries based on multilayered 2D GeAs. The photodetector exhibited a Schottky barrier height (SBH) in the range of 0.40 to 0.49 eV. Additionally, it showed low dark current of 1.8 nA with stable, reproducible, and excellent broadband spectral response from UV to optical communication wavelengths. The highest measured responsivity in the visible is 905 A/W at 660 nm wavelength and 98 A/W for 1064 nm near infrared. Most notably, the planner configuration of this GeAs photodetector showed low detector capacitance below 1.2 pf, low voltage operation (<1V), and a large bandwidth which may exceed 40 GHz. The stability and broadband response of the device are promising for this 2D materials application in high-speed optoelectronic devices.

physics.app-ph↗

arXiv2019

Plasmonic graphene photodetector based on channel photo-thermoelectric effect

We propose an on-chip CMOS compatible graphene plasmonic photodetector based on the photo-thermoelectric effect (PTE) that occurs across an entire homogeneous graphene channel. The proposed photodetector incorporates the long-range dielectric-loaded surface plasmon polariton (LR-DLSPP) waveguide with a metal stripe serving simultaneously as a plasmon supporting metallic material and one of the metal electrodes. Large in-plane component of the transverse magnetic (TM) plasmonic mode can couple efficiently to the graphene causing large temperature increases across an entire graphene channel with a maximum located at the metal stripe edge. As a result, the electronic temperatures exceeding 12000K at input power of only a few tens of μW can be obtained at the telecom wavelength of 1550nm. Even with limitations such as the melting temperature of graphene (T= 4510 K), a responsivity exceeding at least 200 A/W is achievable at telecom wavelength of 1550 nm. It is also shown that under certain operation conditions, the PTE channel photocurrent can be isolated from photovoltaic and p-n junction PTE contributions providing an efficient way for optimizing the overall photodetector performance.

physics.app-ph↗

arXiv2019

Toward Physically Unclonable Functions from Plasmonics-Enhanced Silicon Disc Resonators

The omnipresent digitalization trend has enabled a number of related malicious activities, ranging from data theft to disruption of businesses, counterfeiting of devices, and identity fraud, among others. Hence, it is essential to implement security schemes and to ensure the reliability and trustworthiness of electronic circuits. Toward this end, the concept of physically unclonable functions (PUFs) has been established at the beginning of the 21st century. However, most PUFs have eventually, at least partially, fallen short of their promises, which are unpredictability, unclonability, uniqueness, reproducibility, and tamper resilience. That is because most PUFs directly utilize the underlying microelectronics, but that intrinsic randomness can be limited and may thus be predicted, especially by machine learning. Optical PUFs, in contrast, are still considered as promising---they can derive strong, hard-to-predict randomness independently from microelectronics, by using some kind of "optical token." Here we propose a novel concept for plasmonics-enhanced optical PUFs, or peo-PUFs in short. For the first time, we leverage two highly nonlinear phenomena in conjunction by construction: (i) light propagation in a silicon disk resonator, and (ii) surface plasmons arising from nanoparticles arranged randomly on top of the resonator. We elaborate on the physical phenomena, provide simulation results, and conduct a security analysis of peo- PUFs for secure key generation and authentication. This study highlights the good potential of peo-PUFs, and our future work is to focus on fabrication and characterization of such PUFs.

physics.app-ph↗

arXiv2019

High bandwidth waveguide-integrated plasmonic germanium photodetector

Here we propose a waveguide-integrated germanium plasmonic photodetector that is based on a long-range dielectric-loaded surface plasmon polariton waveguide configuration. As this configuration ensures a long propagation distance, i.e., small absorption into metal, and a good mode field confinement, i.e., high interaction of the electric field with a germanium material, it is perfect for a realization of plasmonic germanium photodetectors. Such a photodetector even without optimization provides a responsivity exceeding 1 A/W for both wavelengths of 1310 nm and 1550 nm. To achieve such a responsivity, only a 5 μm-long waveguide is required for 1310 nm and 30 μm-long for 1550 nm. With optimization this value can be highly improved. In the proposed arrangement a metal stripe simultaneously supports a propagating mode and serves as one of the electrodes, while the second electrode is located a short distance from the waveguide. As a propagating mode is tightly confined to the germanium ridge, the external electrode can be placed very close to the waveguide without disturbing it. As such, the distance between electrodes can be smaller than 350 nm which allows one to achieve a bandwidth exceeding 100 GHz. However, as most of the carriers are generated inside a distance of 100 nm from a stripe, a bandwidth exceeding 150 GHz can be achieved for a bias voltage of -4 V.

physics.app-ph↗

arXiv2019

CMOS-compatible titanium nitride for on-chip plasmonic Schottky photodetector

Here, we propose titanium nitride (TiN) as an alternative plasmonic material for an on-chip silicon plasmonic Schottky photodetector that is based on an internal photoemission process and operating at telecom wavelengths. The examined structure employs an asymmetric metal-semiconductor-metal waveguide structure with one of the electrodes being gold and the second either gold, titanium or titanium nitride. Apart from the excellent optical properties desired for this type of photodetector such as high absorption losses and reasonably high real part of the permittivity, titanium nitride is a CMOS-compatible material that enables easy integration with existing CMOS technology. For the first time, we find a Schottky barrier height of 0.67 eV for titanium nitride on p-doped silicon, which is very close to the optimal value of 0.69 eV. This value ensures very high signal-to-noise ratio of the photodetector operating at a wavelength of 1550 nm. Additionally, TiN provides shorter penetration depth of the mode into metal compared to Ti, which enhances transmission probability of hot electrons to a semiconductor and gives rise to responsivity enhancement.

physics.app-ph↗