Periodic Reporting for period 1 - QUONDENSATE (QUantum reservoir cOmputing based on eNgineered DEfect NetworkS in trAnsition meTal dichalcogEnides)
Reporting period: 2024-04-01 to 2025-03-31
QUONDENSATE is a high-risk/high-gain project with a well-defined and quantifiable goal: to disrupt the state of the art of quantum computing by achieving the first proof-of-concept of quantum reservoir computing (QRC) based on engineered defect networks in transition metal dichalcogenides (TMDs) and demonstrate the advantages of this approach over classical neural networks. Its main objectives are:
O1: Engineering of defects (starting from the pristine ones) and networks thereof by ions/protons irradiation or surface electrochemistry.
O2: Insertion of the TMDs within microcavities to enhance the mutual interaction among the defects through light-matter coupling.
O3: Use of computational approaches to shed light on a) the optical spectra of single defects and networks thereof, b) light-matter interactions in microcavities and ultrafast phenomena, and c) quantum reservoir.
O4: Multiscale physical quantitative characterization of the defect networks. This includes the implementation of a novel noise-correlation spectroscopy and microscopy to analyze pristine defects in chemical vapour deposition (CVD) TMDs on mm2 areas.
O5: Implementation of actual QRC functionalities such as novel machine learning algorithms using a quantum defect network to perform tasks such as image recognition, time series prediction and emulation.
O6: Exploitation/validation of the experimental results done by a key player in the development of photonic quantum computing, Quandela.
O1: Engineering of defects (starting from the pristine ones) and networks thereof by ions/protons irradiation or surface electrochemistry.
O2: Insertion of the TMDs within microcavities to enhance the mutual interaction among the defects through light-matter coupling.
O3: Use of computational approaches to shed light on a) the optical spectra of single defects and networks thereof, b) light-matter interactions in microcavities and ultrafast phenomena, and c) quantum reservoir.
O4: Multiscale physical quantitative characterization of the defect networks. This includes the implementation of a novel noise-correlation spectroscopy and microscopy to analyze pristine defects in chemical vapour deposition (CVD) TMDs on mm2 areas.
O5: Implementation of actual QRC functionalities such as novel machine learning algorithms using a quantum defect network to perform tasks such as image recognition, time series prediction and emulation.
O6: Exploitation/validation of the experimental results done by a key player in the development of photonic quantum computing, Quandela.
During the first year of the project, the consortium has set the stage for achieving its ambitious grand goals by concentrating mainly on O1, O3 and O4. In the following, the progress towards the achievement of these objectives is summarized.
O1: IIT has developed experimental approaches for the synthesis of high-quality monolayer (ML) WS2 by CVD, from isolated single crystals to almost continuous films with size up to wafer scale (2-inch). These samples have been distributed to other project partners for further processing and characterization. UNIMIB has used high-resolution transmission electron microscopy (HRTEM) for the characterization of the defects in the pristine samples and identified single-crystal WS2 grown on pristine sapphire as the least defective material, as compared to a continuous WS2 film or WS2 grown on Al-rich sapphire. This is an important result, as it identifies the best substrate for further processing. INRS has used argon ion bombardment under high vacuum conditions for the controlled generation of defects in ML WS2. X-ray Photoelectron Spectroscopy (XPS) was then employed to characterize the defect density as a function of the fluence of the bombarding ions. A linear relationship between ion fluence and defect density was found. Further methods for defects generation such as electrochemical and proton irradiation-based approaches will be applied in the remainder of the project.
O2: while activities on this objective are scheduled to start later in the project, partner UoW initiated preliminary work on the preparation of the microcavities. Distributed Bragg Reflectors (DBRs) that define the photonic environment for the defects have been designed with the suitable central wavelength (605 nm) and bandwidth (100 nm), in order to match the resonances of defects in WS2.
O3: UU has developed numerical approaches based on density functional theory (DFT) to model point defects (both vacancies and dislocations) in ML WS2, using the Quantum Espresso (QE) code. In particular, both sulphur and tungsten vacancies have been considered and their impact on the electronic band structure and non-resonant Raman spectra has been calculated. Comparison with experimental results achieved by POLIMI and IIT is in progress. CTP PAS has performed preliminary numerical simulations of the reservoir network consisting of a network of defects in the TMD, modelling each network node (defect) as a two-level system. The defects were considered to have slightly different transition energies and decay rates and coherent coupling between nodes and linear optical coupling were assumed. The behaviour of the network was modelled upon coherent laser excitation with different values of the coupling parameter.
O4: POLIMI has implemented a widefield hyperspectral photoluminescence (PL) microscope for characterization of ML TMDs. The microscope, based on an architecture invented by POLIMI, enables acquiring spectral PL hypercubes over wide fields of view with high signal-to-noise ratio. Imaging of samples provided by IIT enabled the identification of localized metallic states introduced by the sulfur-rich zigzag terminated edges and of sulfur vacancies along the bisectors of the WS2 triangles, which are active nucleation sites.
O5: in view of this objective, to be reached at later stages of the project, UoW has developed a setup for time-resolved single-photon counting measurements, allowing the determination of the second-order correlation function, g(2). This capability is critical for assessing whether individual defects behave as isolated two-level systems and for investigating their coupling behaviour to cavity photons and adjacent defects.
O6: QUANDELA, a world-leading player in the field of photonic quantum computing, is responsible for leading the exploitation activities of the QUONDENSATE project. QUANDELA has prepared the Exploitation and Innovation Plan, which describes the Key Exploitable Results (KERs) for each partner as well as the broader impact of the project in the field of quantum computing.
O1: IIT has developed experimental approaches for the synthesis of high-quality monolayer (ML) WS2 by CVD, from isolated single crystals to almost continuous films with size up to wafer scale (2-inch). These samples have been distributed to other project partners for further processing and characterization. UNIMIB has used high-resolution transmission electron microscopy (HRTEM) for the characterization of the defects in the pristine samples and identified single-crystal WS2 grown on pristine sapphire as the least defective material, as compared to a continuous WS2 film or WS2 grown on Al-rich sapphire. This is an important result, as it identifies the best substrate for further processing. INRS has used argon ion bombardment under high vacuum conditions for the controlled generation of defects in ML WS2. X-ray Photoelectron Spectroscopy (XPS) was then employed to characterize the defect density as a function of the fluence of the bombarding ions. A linear relationship between ion fluence and defect density was found. Further methods for defects generation such as electrochemical and proton irradiation-based approaches will be applied in the remainder of the project.
O2: while activities on this objective are scheduled to start later in the project, partner UoW initiated preliminary work on the preparation of the microcavities. Distributed Bragg Reflectors (DBRs) that define the photonic environment for the defects have been designed with the suitable central wavelength (605 nm) and bandwidth (100 nm), in order to match the resonances of defects in WS2.
O3: UU has developed numerical approaches based on density functional theory (DFT) to model point defects (both vacancies and dislocations) in ML WS2, using the Quantum Espresso (QE) code. In particular, both sulphur and tungsten vacancies have been considered and their impact on the electronic band structure and non-resonant Raman spectra has been calculated. Comparison with experimental results achieved by POLIMI and IIT is in progress. CTP PAS has performed preliminary numerical simulations of the reservoir network consisting of a network of defects in the TMD, modelling each network node (defect) as a two-level system. The defects were considered to have slightly different transition energies and decay rates and coherent coupling between nodes and linear optical coupling were assumed. The behaviour of the network was modelled upon coherent laser excitation with different values of the coupling parameter.
O4: POLIMI has implemented a widefield hyperspectral photoluminescence (PL) microscope for characterization of ML TMDs. The microscope, based on an architecture invented by POLIMI, enables acquiring spectral PL hypercubes over wide fields of view with high signal-to-noise ratio. Imaging of samples provided by IIT enabled the identification of localized metallic states introduced by the sulfur-rich zigzag terminated edges and of sulfur vacancies along the bisectors of the WS2 triangles, which are active nucleation sites.
O5: in view of this objective, to be reached at later stages of the project, UoW has developed a setup for time-resolved single-photon counting measurements, allowing the determination of the second-order correlation function, g(2). This capability is critical for assessing whether individual defects behave as isolated two-level systems and for investigating their coupling behaviour to cavity photons and adjacent defects.
O6: QUANDELA, a world-leading player in the field of photonic quantum computing, is responsible for leading the exploitation activities of the QUONDENSATE project. QUANDELA has prepared the Exploitation and Innovation Plan, which describes the Key Exploitable Results (KERs) for each partner as well as the broader impact of the project in the field of quantum computing.
The synthesis of high-quality monolayer WS2 by chemical vapour deposition has been demonstrated, from isolated single crystals to almost continuous films with size up to wafer scale.
Characterization of the defects in pristine samples has been performed using high resolution transmission electron microscopy.
Ad hoc numerical approaches to model the electronic and optical properties of TMDs and their defects have been developed.
Argon ion bombardment under high vacuum conditions has been used for the controlled generation of defects in monolayer WS2. X-ray Photoelectron Spectroscopy was employed to characterize the defect density finding a linear relationship with ion fluence.
A widefield hyperspectral photoluminescence microscope using an innovative architecture has been implemented for the optical characterization of defects in TMDs. Imaging of samples allowed correlating the defects distribution with the growth conditions.
Characterization of the defects in pristine samples has been performed using high resolution transmission electron microscopy.
Ad hoc numerical approaches to model the electronic and optical properties of TMDs and their defects have been developed.
Argon ion bombardment under high vacuum conditions has been used for the controlled generation of defects in monolayer WS2. X-ray Photoelectron Spectroscopy was employed to characterize the defect density finding a linear relationship with ion fluence.
A widefield hyperspectral photoluminescence microscope using an innovative architecture has been implemented for the optical characterization of defects in TMDs. Imaging of samples allowed correlating the defects distribution with the growth conditions.