Nanodevices Research Group

Le metalenti rappresentano oggi una delle frontiere più avanzate dell’ottica moderna: dispositivi ultrapiatti composti da matrici di nanostrutture sub-lunghezza d’onda capaci di modulare fase, ampiezza e polarizzazione della luce con un livello di controllo irraggiungibile per l’ottica tradizionale. Nell’ambito delle applicazioni in crescita nel visibile – dalla microscopia compatta allo sviluppo di sistemi fotonici per il quantum computing – la scelta dei materiali e l’affinamento dei processi di nanofabbricazione risultano decisivi. La presente tesi si propone di affrontare in modo completo lo studio e la realizzazione di una metalente operante tra 600- 800 nm, lunghezza d’onda di particolare rilevanza per le tecnologie quantistiche.
La prima parte sperimentale sarà dedicata alla selezione e alla caratterizzazione dei materiali, concentrandosi su ossido di silicio (SiOₓ) e nitruro di silicio (Si₃N₄), depositati tramite sputtering reattivo. Attraverso spettroscopia ellissometrica verranno determinati spessore, indice di rifrazione composizione nelle condizioni di deposizione, con l’obiettivo di individuare il materiale più idoneo per ottenere meta-atomi ottimizzati per il disegno della metalente. La messa a punto del deposito prevede l’utilizzo dello sputtering presso ILDFA, elissometria, RBS (Rutherfordford Backscattering).
Il progetto prevede anche la modellazione elettromagnetica dei meta-atomi mediante simulazioni FEM, allo scopo di costruire una libreria di nanostrutture in grado di fornire la ritardanza π necessaria per la generazione di fase geometrica. Ciò permetterà di definire la distribuzione spaziale degli elementi per la realizzazione di una q-plate con numero topologico q = 1, capace di generare fasci vortice con momento angolare ℓ = ±2 in funzione della polarizzazione incidente.
La parte centrale del lavoro sarà dedicata alla fabbricazione della metalente. In una prima fase verrà realizzato un prototipo tramite electron beam lithography (EBL), tecnica che garantisce la risoluzione necessaria a definire nanostrutture con dimensioni inferiori ai 100 nm. Seguirà lo sviluppo del processo di trasferimento del pattern nel film dielettrico tramite maschere dure in Al₂O₃ e incisione ICP-RIE. Parallelamente sarà esplorato l’impiego della UV-nanoimprint lithography (UV-NIL), con la realizzazione di master primari e secondari, al fine di valutare una possibile strada per la replica ad alta produttività delle metalenti.
Infine, la tesi prevede la caratterizzazione ottica dei dispositivi ottenuti mediante un banco ottico dedicato, includendo misure di trasmissione co- e cross-polarizzata, ricostruzione della ritardanza spaziale e osservazione dei profili di intensità nel campo lontano per verificare la formazione del fascio vortice.
Il risultato atteso è la dimostrazione sperimentale di una metalente funzionale nel visibile, fabbricata in SiOₓ e capace di generare un fascio vortice allineato con il progetto numerico. Il lavoro offrirà una visione completa – dalla scelta del materiale alla fabbricazione, fino alla validazione funzionale – di un processo tecnologico di grande rilevanza per l’ottica integrata di nuova generazione.

Advancing quantum technologies requires the ability to prepare, manipulate, and measure quantum states in increasingly large Hilbert spaces. One effective approach is to exploit specific degrees of freedom of d-level quantum system to encode qudit states. Qudits offer several advantages over qubits, including enhanced noise resilience, increased channel capacity in quantum communication, improved performance in quantum simulation and machine learning protocols. Photonics platforms are particularly well suited for realizing such states, especially by harnessing the orbital angular momentum (OAM) of light.  When OAM is combined with polarization, forming vector vortex beams, it becomes possible to encode and manipulate high-dimensional quantum states using a single photon, significantly expanding the accessible Hilbert space .
This approach has already been successfully demonstrated in the context of quantum machine learning, where these platforms have enabled effective implementations of quantum reservoir protocols [1,2]. However, scaling to higher-dimensional systems imposes demanding requirements on the control and measurement of quantum states, where each additional dimension corresponds to a distinct spatial mode of the optical field. A major challenge lies in performing mode-resolved measurements that can efficiently discriminate among many spatial and polarization modes. As the number of modes increases, a scalable and effective detection strategy involves combining mode- sorting devices, to spatially separate the modes, with single-photon-sensitive cameras with thousands of pixels, capable of recording multiple spatial channels in parallel.  Such a platform is essential for reading out the evolution of complex photonic qudit states across many dimensions and degrees of freedom. This motivates the development of a metasurface-based mode sorter specifically designed for this spectral window.
Within this framework, this master thesis focuses on the study of random walks of vector beams in both classical and quantum optical regimes, using optical modes carrying orbital angular momentum (OAM). (Fig 1).
The project includes the design, fabrication, and optical characterization of polarization-sensitive metalenses capable of resolving both the topological charge and the polarization structure of the incoming optical field [3,4]. (Fig. 2)
The experimental platform will be used to analyze the evolution of structured beams through a controlled random-walk network and to measure the resulting OAM distributions at the single- photon level.
In a second stage, the same approach will be extended to entangled photon pairs. By constraining their propagation channels through specific OAM spectra, the experiment will enable a test of the Bell inequality under structured-mode conditions. This will allow the verification of quantum correlations in high-dimensional vector spaces and provide insight into the robustness of entanglement under modal scattering.

Fig. 1 Experimental QELM. (a) Schematic overview of the experimental QELM. Initial quantum states |𝜓1⟩,|𝜓2⟩,…,|𝜓𝑛⟩ encoded in the polarization degree of freedom of single photons evolve through a reservoir dynamics, in which the information encoded in the initial two-dimensional space is transferred into the larger Hilbert space of the OAM. By performing only projective measurements on the OAM computational basis, the QELM is trained to reconstruct a set of target values 𝑦1, 𝑦2, …,𝑦𝑛 .

Fig. 2 When the designed metasurface is illuminated by a linearly polarized Gaussian beam, an AV-PVB is formed at the focal plane, with its phase and polarization characteristics determined by the metasurface design. The perfect vortex is generated by leveraging the bifunctionality of the metaoptics, which imparts distinct phase patterns based on the spin of the circularly polarized light in input.

In recent decades, the control of the spatial structure of electromagnetic waves has boosted the research of new modulation formats for telecommunications and quantum applications. In particular, the exploitation of orthogonal spatial modes as distinct information carriers in the so-called mode-division multiplexing (MDM) has led to promising solutions for increasing the information capacity in an efficient and sustainable way. At the single-photon regime, the same techniques allow increasing the number of bits per photon and the security and robustness of quantum communication protocols. So far, MDM has been dominated by beams carrying orbital angular momentum (OAM). However, practical solutions based on OAM-mode multiplexing still suffer from critical issues related, for instance, to efficient generation and long-distance transmissions in free space.
In a more general approach to wavefront propagation, an innovative framework has been recently introduced based on beams with harmonic phases characterized by multipole structures devoid of phase singularities, referred to as multipole-phase beams [1, 2]. This new paradigm offers efficient multiplexing and sorting in a full-optical and compact architecture based on optical transformations, promising, moreover, to overcome the limitations of previous solutions in free-space transmission. Moreover, the multipole order provides an additional degree of freedom, opening to an infinite number of distinct families of multipole phase beams.
In the present work, the student will design and simulate all-optical architectures for the controlled generation and superposition (multiplexing) of multipole-phase beams and their sorting and decoding (demultiplexing) after free-space propagation. In particular, the optical cores of the devices will be fabricated as silicon metasurfaces, leveraging the potential of manipulating light at the nanoscale and the inherent ability to encode more functionalities into the same optical platform. The designed devices will be assembled and tested in the telecom infrared to analyze transmission performance and investigate their use in a real-world scenario.
[1] G. Ruffato, V. Grillo, and F. Romanato, “Multipole-phase division multiplexing”, Opt. Express 29 (23), 38095-38108 (2021)
[2] M. Ferrari, F. Romanato, and G. Ruffato, “Design of a free-space optical link based on multipole-phase division multiplexing”, Opt. Express 33, 4957-4977 (2025)