Doping at the nanoscale: new challenges and advanced applications

Resume : Several novel Chemical Vapor Deposition precursors for group-IV semiconductor doping were introduced in recent years. The common characteristic of these precursors is that they lend themselves to in situ doping of films at low growth temperatures, which is of particular significance for selective growth on device structures and for the growth of group-IV alloys containing Sn. Experiments indicate a very high activation level even without any post-growth annealing step. For some of these precursors, such as trisilyl phosphine, trigermyl phosphine and trigermyl arsine, the substitutional incorporation of the dopant atom is presumably promoted by the fact that the precursor molecule itself incorporates dopant-host atomic bonds. Epitaxial Ge films doped with these precursors display record-low resistivities and dramatically enhanced photoluminescence. Furthermore, the possibility of attaining exceptionally flat doping profiles with very abrupt edges has enabled significant advances in our understanding of the physics of doping in group-IV materials. In this presentation we review a few of these advances, as follows: 1. Incomplete donor ionization in Ge was studied in detail. The results show that the observation of this basic prediction from elementary theory is almost completely frustrated by the merging of the conduction band and the impurity levels. This is markedly different from the case of Si, in which incomplete ionization, while strongly suppressed, is still clearly observable. 2. Fermi-level singularities in the dielectric function of doped Ge were predicted and detected for the first time. These arise from Pauli-blocking of optical transitions to the occupied states in the conduction band of doped Ge. 3. A study of the doping-dependence of the lattice parameter of Ge for different donors (P, As, Sb) reveals a hitherto hidden universal dependence for both Ge and Si, confirming that in addition to a “size” contribution arising from the different atomic radii of dopant and host, there is a large electronic contribution characterized by absolute deformation potentials. 4. Detailed studies of band gap renormalization, including indirect, direct, and higher transitions, were carried out in doped n-type Ge films. The results show that the bands do not shift rigidly as a function of the doping concentration, and that the renormalization depends on the donor. Some of these advances have implications for device modeling. In particular, the need to use degenerate case statistics for highly-doped layers in pin devices is discussed in the context of the observed absence of incomplete ionization phenomena.

Resume : The controlled insertion of electronic states within the band gap of semiconductor nanocrystals (NCs) is a powerful tool which enables to engineer their physical properties. One compelling example is represented by metal chalcogenide NCs incorporating heterovalent p-type impurities such as d10 coinage metals (Cu+, Ag+, or Au+). In this type of nanostructures the interplay between the dopant and the host semiconductor energy levels leads in the ultra-fast capture of the hole by the localized state of the dopant leading to the so-called ?acceptor-bound? excitons. The control of this paradigm unlocked technologically relevant functionalities, such as Stokes-shift between the emission and the absorptionspectra, extended luminescence lifetimes, photomagnetic behaviors, and enhanced electrical transport. To date, although conceptually analogous to hole-management schemes, the opposite ?donor-bound? exciton scheme using aliovalent elements adopted to n-doped NCs (e.g., Al3+ and In3+ in) has not been realized due to the natural propensity of such cations to produce shallow donor states that inject electrons directly in the CB. Here, we exploit the propensity of metal sulfides to present sulfur vacancies (VS) that introduce a localized level pinned about 1 eV below the CB to produce a model system for ?donor-bound? excitons in CdSeS NCs. The investigation of the optical and magneto-optical properties of these NCs revealed that the VS state is responsible for the ultrafast capture of electron in the CB which can then either decay non-radiatively or recombine with the VB photohole leading to long-lived, Stokes-shifted emission with size-tunable energy. Moreover, VS-localized electrons are almost unaffected by trapping, and suppression of thermal quenching boosts the emission efficiency to 85%. Magneto-optical measurements indicate that the VS are not magnetically coupled to the NC bands and that the polarization properties are determined by the spin of the valence-band photo-hole, whose spin flip is massively slowed down due to suppressed exchange interaction with the donor localized electron.

Resume : Laser-assisted Atom Probe Tomography (La-APT) has been applied since over one decade to the analysis of semiconductor systems. Its capability of reconstructing the distribution of chemical species, atom by atom and with spatial precision close to the lattice parameter makes it the ideal technique for the study of problems such as the composition of semiconductor alloys, ordering/disorder in alloy or solute impurity systems, assessment of doping densities and impurity segregation at defects, characterization of interfaces between different crystallographic or compositional phases. This contribution will review the main achievements of the technique in the domain of the characterization of doping behavior in semiconductors. Through exemplary cases, it will be shown how under which conditions a dopant can be detected, how accurate and how precise the determination of doping concentrations can be [1]. Advantages and shortcomings of APT in this domain will be critically addressed, along with the main perspective fields of investigation [2]. Finally, it will be shown how the application of the recently developed Photonic Atom Probe (PAP, a technique allowing for simultaneous APT and micro-photoluminescence spectroscopy) may provide original additional information in the case where the doping is accompanied by specific optical emission properties [3,4]. [1] L. Mancini, et al. The Journal of Physical Chemistry C, 118(41), 24136-24151. [2] E. Di Russo et al. Submitted. [3] J. Houard et al. Review of Scientific Instruments, 91(8), 083704. [4] A. Diaz Damian et al., Application of the photonic atom probe to the study of thick III-Nitride semiconductor heterostructures, submitted to Symposium G.

Resume : The secondary ion mass spectrometry (SIMS) technique has been used to quantify dopant distribution in electronic materials for decades. However, the classical approach, one-dimensional depth profiling, cannot de directly applied to analyze non-planar transistors like fin field-effect transistors and the gate-all-around field-effect transistors (GAAFET). Even though many SIMS instruments are equipped with position-sensitive detectors, the ion bombardment of the three-dimensional structure leads to non-uniform and impossible to predict sputtering process. In this work, we show how to perform SIMS measurements on silicon nanowires (NWs) which can be further processed into GAAFET devices. Before the measurement, the structure is embedded in an organic matrix to ensure that NWs are sputtered from the top and not from aside. However, for standard measurement conditions, the sputtering rate of the organic material is more than an order of magnitude higher than that of the silicon which impedes the analysis. Application of high incident-angle ion bombardment practically eliminates this difference and a sample can be uniformly sputtered and boron distribution along NWs can be quantified. The analysis has been performed on an array of 100 x 100 boron-doped silicon NWs which were subjected to various oxidation processes followed by oxide removal. SIMS analysis confirms the segregation of boron towards the oxide and its depletion from the cores of nanowires. The results are in line with TCAD 3D simulations. Moreover, the intensity of the silicon signal scales linearly with the area of NWs and thus their diameter can be determined directly from the SIMS measurement, which proves the stability of the proposed measurement procedure. Acknowledgments: The research leading to these results has received funding from the European Union's Horizon 2020 research and innovation program under grant agreement No. 871813 MUNDFAB and the National Centre for Research and Development under grant agreement No. LIDER/8/0055/L-12/20/NCBR/2021

Resume : The ultra-fast heating/melting provided by the pulsed Laser Annealing (LA) process is an important resource for the manufacturing of conventional and quantum devices. The localization of the annealing/melting in ns/nm time/space scales permits the optimal application in the fabrication flow of: power devices, CMOS logic and 3D sequentially integrated devices, front end and back end structures, memories, hyper-doped and ultra-doped opto- and quantum- devices, etc. (see Ref. [1] for an overview of the current technological approaches). For many of the cited applications the complete control of pulsed LA process is still critical due to the presence of 2D/3D nanoscale space distributions of different materials where the irradiation effects are strongly non-linear. As a consequence, any demonstration of feasible process integration needs a complete retraining also for limited design change of the device and the predictive simulation is a key element for the definitive qualification of LA in nano-device manufacturing. In this contribution a review of the current status of LA modelling will be presented with a particular focus on the need of atomistic approaches. Indeed, the correct prediction of physical phenomena occurring during LA at the nanoscale, such as defect or stress generation and evolution, explosive recrystallization and shape deformations, requires atomistic resolution. In particular, we present a multi-scale simulation methodology based on an algorithm that couples a finite-elements model for macroscale e.m. field and temperature evolution in the device/environment with a Super-Lattice Kinetic Monte Carlo scheme, able to model with atomic resolution the highly crystal-orientation dependent evolution of liquid-solid interfaces within critical nanoscale portions of the device [2]. Self-consistent kinetic simulations clearly demonstrate that the propagation of the laser electromagnetic field, its absorption and all the phenomena activated by the structure-e.m. field interaction depend critically on the complex geometry and materials design of the devices. Several difficult challenges emerge when considering possible LA applications and they will be discussed along with experimental analyses. Finally, we also indicate the guidelines for the future computational research in the field; and, in particular, the role of the nano-constrains in the predictions of the temperature evolution during the process. We acknowledge this research project has received funding from the European Union’s Horizon 2020 research and innovation programme under Grant Agreement No. 871813 (MUNDFAB). [1] Laser Annealing Process in Semiconductor Technology: Theory, Modelling and Applications in Nanoelectronics, F. Cristiano and A. La Magna Eds, Elsevier (2021). [2] G. Calogero, D. Raciti, P. Acosta-Alba, F. Cristiano, I. Deretzis, G. Fisicaro, K. Huet, S. Kerdilès, A. Sciuto, A. La Magna, "Multiscale modeling of ultrafast melting phenomena." npj Computational Materials 8, 36 (2022).

Resume : Attaining Ge1-ySny alloys with high Sn content is a keystone for a large number of applications ranging from high performance nanoelectronics to integrated mid-infrared photonics in Si [1]. Here, we present a novel approach for the fabrication of fully relaxed Ge1-ySny layers on Ge with Sn fraction up to 13 % and very high crystalline quality. The incorporation of Sn in Ge was obtained by sputtering of thin Sn films (< 20 nm) directly on Ge wafers followed by laser pulsed melting that leads to the diffusion of the Sn in Ge [2]. The concentration of Sn in the alloys was varied as a function of the thickness of the Sn film and the laser process parameters (number of shots). Microstructural analyses combining high-resolution transmission electron microscopy, atom probe tomography and nanobeam precession electron diffraction were performed to investigate the Sn distribution and the strain state down to the nanoscale. Ge1-ySny layers with y > 6 % are fully-relaxed with respect to the Ge substrate, and Sn-rich regions are formed in correspondence of dislocations. With the exception of these regions, Ge1-ySny alloys present a very homogeneous and random Sn distribution, with all Sn atoms located in substitutional positions, as revealed by Rutherford back-scattering measurements. The new approach adopted in this work offers an attractive alternative to epitaxy or ion implantation to locally fabricate high quality Ge1-ySny alloys, with possible attractive developments for the production of direct bandgap Ge-based alloys by adopting strain engineering techniques. [1] S. Wirths, D. Buca, S. Mantl, Si-Ge-Sn alloys: From growth to applications, Prog. Cryst. Growth Charact. Mater. 62 (2016) 1?39. [2] C. Carraro, et al., N-type heavy doping with ultralow resistivity in Ge by Sb deposition and pulsed laser melting. Applied Surface Science 509 (2020): 145229.

Resume : An effective doping technology for shallow junction formation using P atom injected into a semiconductor substrate by pulsed laser annealing is presented. Polystyrene polymers with a narrow molecular weight distribution and end-terminated with a P containing moiety are used to build up a phosphorus δ-layer on deglazed and not deglazed high resistivity (100) Si substrates. The P δ-layer is used as dopant source. P atoms are injected into the Si substrate by pulsed laser melting. (PLM). No P diffusion is observed in the case of not deglazed silicon surfaces, indicating that the thin SiO2 film that is naturally formed on the Si surface when exposed to air represents a strong diffusion barrier for P atoms during PLM. Conversely, Efficient incorporation of P atoms into the substrate is achieved in the case of deglazed Si substrates forming shallow junctions with a box like distribution of P atoms. The possibility to modulate P distributions into the deglazed Si substrate is investigated by changing the energy and number of the laser pulses. P concentrations higher than 1 x 1019 atoms/cm3 are obtained and the depth of the junction is varied from 20 to 100 nm. Additionally strategies to minimize carbon and oxygen contaminations during the PLM process are developed. The use of the PLM technique allows achieving a shallow and homogeneous doped region with a high concentration level of phosphorus atoms.

Resume : Ge1-xSnx alloys have attracted considerable attention for their promising electrical and optical properties. One of the main challenges for their successful implementation in devices concerns the fabrication of n-type heavily doped surface layers. In this presentation, a new methodology for ex-situ doping of Ge1-xSnx layers is presented. It consists in the deposition of Sb atoms on the surface of Ge1-xSnx layers followed by pulsed laser melting (PLM) that ensures the diffusion of Sb into the alloy. We demonstrate that Sb is incorporated very efficiently within a relaxed Ge0.91Sn0.09 epilayer, with supersaturated 4 × 10^(20) cm^(-3) active concentrations, in line with literature records obtained in Ge1-xSnx with in-situ approaches. At the same time, we evidence that the concentration of substitutional Sn close to the surface decreases from 9 to about 6 at. % after PLM, inducing a contraction of the lattice parameter perpendicular to the underlying Ge1-xSnx. These results demonstrate a possible route for ex-situ n-type heavy doping of Ge1-xSnx alloys but indicate also that Sn redistribution and precipitation phenomena need to be carefully considered for a successful process development.

Resume : Nanosecond laser annealing (NLA) might improve the performances of GeSn-on-Ge light emitting devices, through improvements of their structural properties and dopant activation. Compared to mainstream annealing processes, it is so short (less than a few hundreds of nanoseconds, typically) that Sn segregation can be prevented. Therefore, NLA offers a unique opportunity to anneal GeSn in a tightly controlled, far from equilibrium way. Pulsed NLA experiments were thus conducted on tens of nm thick (i) intrinsic GeSn 6%, 10% and 14% layers and (ii) P implanted (1x1015 cm-2) GeSn 6% layers pseudomorphically grown on Ge-buffered Si substrates. NLA was performed in a SCREEN-LASSE tool with energy densities (ED) as high as 1.6 Jcm-2, a pulse duration of 160 ns and a lasing wavelength of 308 nm. Time Resolved Reflectometry (TRR) measurements were recorded during laser annealing thanks to a 638 nm wavelength laser. Small surface structures formed at the melt threshold (ED = 0.60 Jcm-2 – 0.775 Jcm-2) did not change the TRR signal when multiple laser pulses were used at the same position on intrinsic GeSn layers. At higher EDs, larger surface structures appeared, with a TRR signal reduction, then. A smooth surface was recovered when melting the entire GeSn layer (ED = 1.00 Jcm-2 – 1.025 Jcm-2). The liquid / solid interface was then the smooth GeSn / Ge interface, with the formation, after solidification, of a smooth surface. When more than 10 cumulative laser pulses were used with ED = 1.00 Jcm-2 – 1.025 Jcm-2, the top of the Ge Strain-Relaxed Buffer (SRB) was melted and large surface structures were formed. Secondary Electron Microscopy with Energy Dispersive X-ray spectroscopy hinted at the islands being Sn rich. A shift of the GeSn (0 0 4) X-ray Diffraction (XRD) peak towards that of the Ge SRB occurred for multiple laser pulses. Secondary Ion Mass Spectrometry profiles showed that such a shift was due to Sn redistribution in the stack. As far as phosphorous implanted GeSn is concerned, explosive recrystallization was observed for ED = 0.45 Jcm-2 and a second melt around 0.70 Jcm-2. The TRR tail around 1.10 Jcm-2 was a sign of the smoother surface obtained when fully melting the GeSn:P layer. AFM outlined the impact of the starting crystalline quality on the orientation of surface structures. They were more dispersed, after NLA, for ion-implanted GeSn:P than for intrinsic GeSn. Reciprocal Space Maps and ω-2Θ scans otherwise showed that GeSn:P layers stayed fully compressively strained after NLA, although Sn was redistributed. High quality GeSn:P layers with multiple well-defined XRD peaks and a reduced sheet resistance R_S were obtained for an ED of 1.00 Jcm-2. R_S was as low as 48 Ω/sq after NLA at 1.30 Jcm-2, with part of the Ge SRB melted, then. Solid Phase Epitaxial Recrystallization was otherwise investigated at low EDs and thus reduced temperatures. It might in the future enable dopant activation in ion-implanted GeSn without Sn redistribution.

Resume : Semiconductor properties of two-dimensional (2D) materials associated with the additional degree of freedom originating from magnetic ion spin orientation is subjected of intense investigation due to its promising application in 2D spintronic devices. Mn-doped materials from Transition Metal Dichalcogenides (TMDs) family can exhibit enhanced magnetooptical properties and fascinating magnetic phenomena as e.g. carrier mediated ferromagnetism. Here we report on an extensive study on growing MoSe2 layers mixed with manganese using molecular beam epitaxy. To study impact of Mn ions on properties of TMD we have grown in the same conditions: undoped molybdenum diselenide (MoSe2) and a series of Mn – doped molybdenum diselenide ((Mo,Mn)Se2) samples. The following substrates were used: Si with polycrystalline SiO2 (SiO2/Si) buffer, (SiO2/Si) with exfoliated hexagonal Boron Nitride (hBN/SiO2/Si) flakes on top, and graphite on tantalum foil. On SiO2/Si substrate we have grown a series of samples with various amount of deposited Mn, while amount of Mo and Se were kept constant (optimized for 1 monolayer of MoSe2). Next, we investigated all samples using room temperature optical spectroscopy: Raman scattering and photoluminescence (PL). We have observed that characteristic Raman line of MoSe2 at 241 cm-1 only weakly evolved with increasing amount of Mn – has slightly shifted towards higher energies. Also, we found that the addition of manganese have not significantly altered the intensity of PL. There are only weak effects of PL shifting, MoSe2 peak on 1.56 eV shifts by approximately 20 meV under influence of Mn ions. Si substrate with 90 nm thick SiO2 buffer is very convenient for optical study of (Mo,Mn)Se2, however such kind of polycrystalline buffer gives no hope for growth of large monocrystalline layers and better incorporation of Mn dops into the MoSe2 layer. This is why we decided to start work with hBN flakes. In this case, we have crossed a phase segregation problem. HBN is the inalienably perfect substrate for MoSe2 [1] growth, but trying to grow (Mo,Mn)Se2 we have found out that this compound divide into MoSe2 and MnSe [2] forming two separate phases, what we have confirmed using PL measurements. For better understanding of atoms distribution in studied compound, we have performed Scanning Transmission Microscopy investigation utilizing samples grown on graphite. Results of STM imaging revealed coexistence of two crystallographic MoSe2 phases: 2H – the most popular and stable semiconductor one and 1T’ – more rare, metallic phase, that can act as Weyl semimetal. [1] Pacuski, W. et al. Narrow Excitonic Lines and Large-Scale Homogeneity of Transition-Metal Dichalcogenide Monolayers Grown by Molecular Beam Epitaxy on Hexagonal Boron Nitride. Nano letters 20, 3058–3066; 10.1021/acs.nanolett.9b04998 (2020). [2] Kucharek, J., Bożek, R. & Pacuski, W. MnSe - Molecular Beam Epitaxy Growth and Optical Characterisation. Acta Phys. Pol. A 136, 598–602; 10.12693/APhysPolA.136.598 (2019).

Resume : The optoelectronic features of metal oxide nanocrystals (MO NCs) are substantially influenced by surface depletion regions caused by surface states. Furthermore, MO NCs show a localized surface plasmon resonance (LSPR), allowing for tunable properties enabled by doping or electrochemical or photochemical charging. MO NCs' dynamic control over the LSPR makes them intriguing for a variety of optoelectronic and storage applications. For all these reasons, the manipulation of the NC depletion width allows for tuning the NC's characteristics. However, the mechanism behind this phenomenon is extremely complicated and not entirely understood. It is possible to engineer the depletion region by modifying different factors such as the material under consideration, the size of the NCs, and the existence of several core-shell systems. To accomplish so, the band and carrier density profiles of NCs may be calculated considering the aforementioned features. In this work, a novel paradigm for predicting the behavior and the physics of the MO NC photodoping process is provided, indicating that the charging mechanism is unexpectedly based on the electrical rearrangement of the energy bands. The case of a core-shell structure of Sn:In2O3/In2O3 NCs was numerically and experimentally studied by modifying the thickness of the shell, as well as post-synthetically, both by photodoping and reversible chemical processes. The engineering of the depletion layer, and the subsequent manipulation of the electronic structure allow for a large improvement in LSPR sensitivity and the targeting of peculiar features in MO NCs. The fine-tuning of the band structure of NCs has resulted in an increase in charge storage capacity, which is a step toward fully light-driven energy storage devices.

Resume : Because of their size and ability to precisely confine a single spin, single dopant atoms in silicon are very logical candidates for a quantum bit (qubit) [1], the building block of a quantum computer. In particular, heavy dopant atoms will provide unprecedented coherence times because their quantum states are extremely well protected from outside interference [2], firstly because of the near-absence of nuclear spins in silicon, and secondly because of the small effective Bohr radius of deep dopant atoms. Firstly, we fabricate nano-MOSFET structures in silicon. We investigate the properties of the materials used for spin qubits in silicon nanostructures, namely the conformity of Al, Ti and Pd nanoscale gates by means of transmission electron microscopy [3]. Next we define low-disorder electron quantum dots with Pd gates [4], which we use as charge sensors capacitively-coupled to a neighbouring dopant atom. Secondly, we aim at the readout and manipulation of individual spins of heavy dopants in silicon. We implement a highly-sensitive active charge sensing technique [5] coupled to time-resolved measurements to enable single-shot readout of individual spins in silicon. We further use spin resonance experiments for single spin qubit control. [1] S. Asaad et al. Nature 579, 205 (2020) [2] G. Wolfowicz et al., Nature Nanotech. 8, 561 (2013) [3] P. C. Spruijtenburg et al., Nanotechnology, 29, 143001 (2018). [4] M. Brauns et al., Scientific Reports 8, 5690 (2018). [5] A. J. Sousa de Almeida et al., Phys. Rev. B 101, 201301(R) (2020)

Resume : Motivation Quantum technologies and solid-state quantum computation based on the intrinsic two-level dynamics of electron spin in semiconductors has attracted widespread attention because of the strong microelectronic integration. Within the subsector of single dopant-based qubits there are two main sets of dopant and substrate which look to be most promising: a single dopant impurity in a silicon substrate [1] and the nitrogen vacancy center in diamond [2]. Single dopants have exhibited extremely long spin-coherence and spin-relaxation time [3] which contribute to the critical ‘closed box’ requirement of a quantum system, and coherent control of impurity wave functions of these single dopants has also been demonstrated [4]. A number of techniques have been utilised to construct solid state quantum devices including the use of scanning tunneling microscopy and hydrogen resist lithography for single atom manipulation [5]. These have demonstrated the potential of such devices but are very limited in their flexibility and ability to scale-up. Current ion beam technology can produce sub 10nm spot sizes with nm precision of beam placement, but subsequent ion implantation creates damage that must be annealed. There are, however, significant advantages in terms of flexibility and vastly greater speed with the ability to implant a range of ion species into substrates enabling levels of scale-up far beyond current laboratory developed devices. We give two aspects where the use of ion implantation can be employed in the processing of solid state quantum devices. 1. Single Ion Implantation SIMPLE- Single Ion Multispecies Positioning at Low Energy, is an ion beam system with the capability to deterministically place ions into a substrate with sub-20nm precision [6]. The tool is being developed at Surrey with Ionoptika and uses current state of the art focused ion beam technology to act as an implantation tool, along with ultrahigh vacuum systems to create a clean environment for the construction of qubit systems. In the regime of controlled single ion implantation, the process is essentially statistical in nature and the key to deterministic implantation is a high (>95%) detection capability for each ion as it arrives at the surface. A series of liquid-metal ion sources capable of delivering a range of ions including P, Si, S, Se, Bi and Er will be available along with a separate instrument fitted with a gas source for N, C and O ions. The current system will be described and its operational parameters explored. Although more than 48 different ion species across the periodic table have been demonstrated in the past [7], it is clear that many of these have less than optimum “lifetimes” and often do not produce a stable beam for prolonged use. An on-going development program between Surrey and Ionoptika on the production of source materials and tip shape and design for improved liquid metal ion source stability and lifetime is underway. The use of a conventional Focussed Ion Beam (FIB) to provide different shaped tips to find the optimum design for a given source material is just one area under exploration here. 2. Silicon Isotopic Enrichment One of the most important aspects of solid-state quantum devices fabricated in silicon is that the donors must be isolated from environmental perturbations by being sited in a defect-free, cryogenically cooled, isotopically pure 28Si crystal which acts as a ‘semiconductor vacuum’. Naturally abundant silicon has a 28Si content of 92.2%. 29Si atoms, with nuclear spin I = 1/2, occur at an abundance of ∼4.7% and can disturb qubit states via nuclear spin interactions. 30Si atoms, which occur at 3.1% abundance, have no spin but their different mass can still cause localised magnetic field variations which hinder uniform nuclear magnetic resonance manipulation across all qubits [8]. Reduction of the level of 29Si and 30Si or, conversely, enrichment of the level of 28Si, is therefore a key requirement for the realisation of silicon-based quantum computing. We have recently [9, 10] demonstrated the use of a conventional ion implantation to enrich a 28Si target to levels required for quantum devices. This has been achieved via two routes: i) direct implantation of an isotopically pure layer into the silicon substrate; and (ii) using a ion implantation into an Al film and using layer exchange to move the implanted, isotopically pure layer, to the surface of the silicon substrate. We will show the advantages and disadvantages of these techniques References [1] B. E. Kane, Nature. 393, 6681 (1998). [2] M. V. G. Dutt et al., Science. 316, 5829 (2007). [3] A. M. Tyryshkin et al., Nat. Materials. 11, 2 (2011). [4] P. T. Greenland et al., Nature. 465, 7301 (2010). [5] M. Fuechsle et al., Nat. Nanotechnology. 7, 4, (2012). [6] N. Cassidy et al., Phys. Status Solidi A, 218, 2000237, (2020) [7] Bischoff, Lothar, et al. Applied Physics Reviews 3.2, 021101, (2016) [8] K.Itoh and H.Watanabe, MRS Communications, 4, 143–157, (2014) [9] E.Schneider, J.England, L.Antwis, A.Royle, R.Webb, R.Gwilliam, J.Phys D., 54, 355105, (2021) [10] J.England, D.Cox, N.Cassidy, B.Mirkhaydarov, A.P Fadon, Nucl Inst. Meth.B, 461, 30, (2019)

Resume : Deterministic doping, namely the capability to control the position of dopant atoms at the nanoscale, is a very coveted goal in semiconductor technology because of its wide range of application: it can be used in standard CMOS technology to improve the performances of devices (1) or to explore new kind of computational architectures, such as solid-state semiconductor quantum computers or natural computation (2). Despite the promising expectations, this level of in depth and lateral control of dopants can be achieved only by energy-consuming serial processes with extremely low throughput. Alternative strategies are required to control positioning of impurity atoms within a semiconductor with nanometer accuracy. Our purpose is to approach this goal creating a modulation of dopant atoms at the nanoscale using a time and energy-saving parallel process that combines a standard top-down technique, ion implantation, with block copolymer (BCP) lithographic masks. In this respect, the application of BCP as lithographic mask for ultra-low energy ion implantation has already been demonstrated (3) with a Si implant. In more details, we explored the possibility to control the positioning of the P impurity atoms in the Si substrate by ultra-low energy implantation through a mesoporous polystyrene (PS) mask. The implantation energy was set at 3 keV and four different fluences were used, ranging from 1.5e+14 cm^-2 to 5.5e+14 cm^-2. The mask was obtained by selectively removing the polymethylmethacrylate (PMMA) phase from a 35 nm thick self-assembled PS-b-PMMA film, deposited on top of a not deglazed Si substrate. The mask consisted of a mesoporous template with out of plane hexagonally packed pores, having a diameter of 20 nm and a center-to-center distance of 35 nm. Accordingly ordered arrays of P doped Si nano-volumes were generated within the Si substrate. SEM and AFM analyses were performed to characterize the BCP templates before and after implantation, as well as to identify the implanted regions in the Si substrate. A local increase of the substrate roughness, matching the pattern of the mask, confirmed the effectiveness of PS film to block P ions. ToF-SIMS depth profiling showed a decrease of the P signal with respect to samples implanted without the PS. The dose reduction was determined to be proportional to the area of the pores in the mask. Operating at implant fluences near the amorphization threshold, we activated the dopants by thermal treatments at relatively low temperatures (500°C-700°C), in order to prevent P diffusion and to preserve their confinement in nano-volumes with characteristic dimensions below 20 nm. The amorphous layer formation upon implantation and its recrystallization after annealing were studied with UV micro-Raman, while ToF-SIMS was performed to study the P diffusion profiles. 1)Shinada et al, Nature, 437, 7062, 1128–1131, 2005 2)T. Chen et al, Nature, 577, 7790, 341–345, 2020 3)C. Castro et al., Nanotechnology, 24, 7, 2013

Resume : An effective doping technology for precise control of P atom injection and activation into a semiconductor substrate is presented. Polystyrene polymers with a narrow molecular weight distribution and end-terminated with a P containing moiety are used to build up a phosphorus δ-layer on deglazed and not deglazed high resistivity (100) Si substrates. The P δ-layer is used as dopant source and P atoms are efficiently injected into the Si substrate by high temperature (900–1250 °C) thermal treatments in a rapid thermal processing (RTP) machine. Depth profiles of the injected phosphorus atoms are obtained by ToF-SIMS analysis of the samples upon annealing, demonstrating the effective capability to modulate the amount of dopants diffused in the Si substrate. Activation rates (ηa) of injected P atoms are investigated as a function of the processing conditions by temperature dependent (100 – 300 K) resistivity and Hall measurements in the van der Pauw configuration. Significantly, different activation rates are obtained when using deglazed and not deglazed Si surface for the grafting of polymers terminated with the P containing moiety. In the case of not deglazed Si surface, activation rates ηa ~ 60% are obtained, while in the case of deglazed Si substrate ηa > 85% are measured, indicating an almost complete activation of the injected dopants. This bottom-up approach holds promise for the development of a mild technology for efficient doping of semiconductors.

Resume : Including impurities in a crystal is a process known as doping. It is of vital importance for our semiconductor technology as it allows to engineer the electronic and optical properties of semiconductors. Such process can be applied to bulk crystals as well as to nanocrystals [1]. As graphene is made up of only carbon atoms [2] , one could envision replacing some of these with either nitrogen or boron, which are first-neighbours in the periodic table, thus doping it. Indeed, many research groups have already reported it [3, 4]. Recently, however, oxygen atoms within a graphene oxide lattice were directly observed using an aberration-corrected scanning transmission electron microscope [5]. Here, we successfully develop a low-energy plasma-based method to include oxygen in graphene [6]. Scanning tunnelling microscopy is used to visualize the oxygen impurities within the graphene lattice and the collected experimental images are compared to theoretically simulated ones. Within our samples, graphitic substitution - that is, when oxygen binds three neighbour carbon atoms - and a pair of oxygen atoms substituting two nearest neighbour carbon atoms are the configurations most likely observed. Furthermore, the electronic properties of these graphene samples with substitutional oxygen are examined. Interestingly, while oxygen bonded to graphene basal plane in the form of functional groups p-type dopes graphene, here we show that substitutional oxygen n-type dopes it. Moreover, transport measurements and atomistic calculations show an asymmetric conduction in doped graphene, with electrons being less scattered than holes. These findings will be beneficial to push further the application of carbon nanomaterials, such as graphene oxide, oxidized carbon nanotubes or novel two-dimensional π-conjugated organic frameworks systems, as they reveal the impact of oxygen in them. References [1] Erwin, S. C. et al. Nature 436, 91 (2005). [2] Novoselov, K.S. et al. Science, 306, 666 (2004) [3] Agnoli, S. et al. J. Mater. Chem. A, 5002 (2016) [4] Wang, H. et al. ACS Catalysis, 2, 781 (2012) [5] Hofer, C. et al. Nature Communications, 10, 4570 (2019) [6] Mackenzie, D. M. A. et al. 2D Materials, 8, 045035 (2021)

Resume : Diamond has unique properties that make it an attractive wide band-gap material to produce future high-performance electronic devices. With a wide band-gap of 5.5eV, a thermal conductivity 5 times greater than 4H-SiC, a high breakdown field and high hole and electron carrier velocities, diamond is a clear stand out candidate for high frequency and high power devices. However, the lack of a suitable doping mechanism has hindered the application of diamond in electronic devices. Conventional substitutional doping techniques are limited as it is difficult to substitute atoms into the diamond crystal lattice, limiting the potential for diamond to be used as a semiconductor in electronic applications. Surface Transfer Doping (STD) gives the use of diamond for such applications more promise. For STD to occur there are typically two prerequisites: hydrogen terminated diamond (H-diamond) and an electron accepting material in intimate contact with the H-diamond surface. The hydrogen termination gives the diamond a negative electron affinity which facilitates the transfer of electrons from the diamond to the electron-accepting material, creating a shallow, quasi two-dimensional hole gas (2DHG) in the diamond. This doping process traditionally relies upon interfacial electron transfer between the diamond valence band and favourable energy states provided by atmospheric molecules dissolved in a water layer naturally adsorbed on the diamond surface. However, the stability of this atmospheric layer, upon which the transfer doping process relies, has been a significant limiting factor in the production of high-power handling and robust operation devices. Materials that can improve the performance and stability of STD in diamond include the metal oxides such as MoO3 and V2O5 which act as an alternative electron acceptor medium on the H-diamond surface. In order to validate and understand the physical and the chemical process in such STD, in this work we have combined experimental and simulation studies. The electrical characterisation is done by high temperature Hall measurements. Those experimental results are compared to numerical simulation based on the first principle methods such as Density Functional Theory (DFT). DFT was used to calculate the band structure and charge transfer process between these oxide materials and hydrogen terminated diamond. Analysis of the band structures, density of states, Mulliken charges, adsorption energies and energy levels shows that both oxides act as electron acceptors and inject holes into the diamond structure. These results demonstrate that the oxygen content of these metal oxides plays a major role in the surface transfer doping of H-diamond.

Resume : This seminar presents a brief summary of our research in boron related materials from boron nitride (BN). BN nanomaterials including nanotubes, nanosheets, nanowires and nanorods will be introduced from synthesis to properties and various applications. BN nanosheets (2D h-BN) will be presented in details. BN nanosheet also known as white graphene, has a similar honeycomb structure to graphene, with alternating boron and nitrogen atoms consisting of strong sp2 covalent in-plane bonding and weak van der Waal forces between layers. h-BN has unique properties; it is a lubricant due to its layered structure. BN is an electrical insulator (with a bandgap of approximately 5.9eV) at the same thermally conductive and also has high mechanical strength, large thermal conductivity and low dielectric constant. We will demonstrate our research experience and extertise in developing new target materials for fusion reaction in different structures and forms from laboratory scale to commercialisation.

Resume : One of the most important challenges concerning the fabrication of devices based on MoS2 is to find novel bottom-up material growth methodologies as alternative to chemical vapor deposition (CVD), which is the most commonly used growth method. Recently, sputtering has gathered a particular attention due to the simplicity of the growth method jointly with its reliability, large area growth possibility and repeatability [1]. However, sputtered layers require post-deposition thermal treatments in order to obtain crystalline MoS2. Recently, pulsed laser annealing (PLA) was emerged as smart processing technique toward large-scale crystals without affecting the underlying substrate [2]. In particular, the UV pulses of nano-second lasers induce ultra-rapid thermal treatments leading to extremely controlled and confined thermal processes. Here, we will introduce the progresses in growing large area and high-quality crystalline MoS2 (layer thickness < 10 nm) on both Si and SiO2 substrates by sputtering deposition followed by PLA using a KrF excimer laser. The evolution of the material crystallinity varying the substrate material, the sputtering deposition conditions and the PLA process parameters was studied by performing Raman spectroscopy and GI-XRD. In addition, the material stoichiometry (Mo and S areal doses) was determined by Rutherford Backscattering Spectrometry (RBS) varying the sputtering/PLA conditions. The results obtained reveal the formation of (002)-oriented MoS2 with good material quality. The processes identified in this work offer an attractive solution to locally fabricate MoS2-based devices on Si-based electronics. [1] Merve A. et al., Sputtered 2D transition metal dichalcogenides: from growth to device applications. Turkish Journal of Physics, 45(3), 131-147 (2021). [2] Sirota B. et al., Room temperature magnetron sputtering and laser annealing of ultrathin MoS2 for flexible transistors. Vacuum, 160, 133-138, (2019).

Resume : Inorganic halides have been extensively exploited in various applications such as lasers, field-effect transistors (FETs), light-emitting diodes (LED), water splitting, water purification and solar cells owing to their exceptional electronic and optical properties. Single wall carbon nanotubes (SWCNTs) with well-known chemical robustness provide an excellent host to encapsulate materials and manufacture structures as thin as a single atom to a few atoms in thickness. The confinement of the structures within SWCNT provides the opportunity of producing materials of dramatically lower coordination number compared to the corresponding bulk materials. Additionally, confining these materials to a few atoms thick in cross section changes their electronic properties due to quantum size effects and these can be measured. We also show how changing the size of the encapsulated material can also contribute to the tuning of the composition. Here we demonstrate stable 1D inorganic halide chains from cesium iodide within the narrowest discrete SWNTs with diameters below ~0.8nm and up to ~1.1nm and report on their structural evolution as the diameter increases. For SWCNTs with somewhat larger average diameters ranging between ~1.2-1.6 nm, we have discovered four new structures encapsulated formed from the same elements employed in the formation of lead and tin based halide perovskites. We have studied the critical role of the size of the host channels in creating and modulating the structure and stoichiometry of the smallest scale halide phases imaginable. We have successfully identified new novel chemically stable halides with the battery of single atom chemical mapping and imaging supported by extensive DFT calculations. Finally, 1D heterostructures offer further tangible benefits. The oxidation or chemical resistance of carbon nanotubes can be improved by the protective BN sheath. A plethora of opportunities arises when stacking 1D crystals coaxially, such as interesting 1D physics in heterostructure electronics. The hybrid shells of 1D heterostructures can be chosen with an electronic band alignment such that interlayer excitons electrically neutral, short-lived particles normally caused by light become favoured even in the ground state, reaching Bose-Einstein condensation into an excitonic superfluid. This enables scientists to build low-power logic devices or direct-current electrical transformers. Coaxially grown 1D CNT/ BN/ MoS2 heterostructure’s structure and optical properties characterised by electron microscopy, Raman Spectroscopy, optical pump-optical probe spectroscopy will be presented.

Resume : To address semiconductor sustainability challenges in the next decades, ZnO nanowires (NWs) are one of the most promising candidates due to their attractive structural, optical, and electrical properties [1]. The control of physical properties using doping has deeply been studied when ZnO NWs are grown by vapor phase deposition techniques and more recently by chemical bath deposition, as a low-cost and low temperature method in aqueous solution [2]. A relevant approach has consisted in adding extra metal salts containing the dopants in the bath and tuning its pH to drive attractive electrostatic forces that are favorable for the incorporation of these dopants into ZnO NWs [3-5]. Such an approach has been applied so far to the incorporation of Al, Ga, and Cu, but its relevance for the development of co-doping as a promising route to further control the physical properties of ZnO NWs is still under debate. In this work, we study the simultaneous co-doping of ZnO NWs with Al and Ga by adding Al(NO3)3 and Ga(NO3)3 in a single bath containing Zn(NO3)2 and HMTA in deionized water, while adjusting its pH using NH3. The physicochemical processes operating in the bath along with the morphology and structural properties of ZnO NWs are investigated in detail using thermodynamic simulations, scanning electron microscopy, and X-ray diffraction. The incorporation of Al and Ga dopants is assessed by X-ray photoelectron spectroscopy, Raman spectroscopy in low and high wavenumber ranges, and inductively coupled plasma mass spectrometry. It is shown that ZnO nanowires can simultaneously be co-doped with Al and Ga, but that the incorporation processes are correlated each other and affect the residual doping arising from hydrogen. The present investigation casts a light on the co-doping of ZnO NWs grown by chemical bath deposition, opening some perspectives to design multi-doped ZnO NWs for nanoscale engineering devices. [1] Y. Zhang et al. J. Nanomater., 2012, 2012, 1‑22. [2] S. Xu et al. Nano Res., 2011, 4, 1013-1098. [3] C Verrier et al. Inorg. Chem. 2017, 56, 13111-13122. [4] P Gaffuri et al. Inorg. Chem. 2019, 58, 10269−10279. [5] C Lausecker et al. Inorg. Chem. 2021, 60, 1612−1623.

Resume : Creating atomically thin nanowire (NW)-filled single-walled carbon nanotubes (SWCNTs) enables the study of unconventional charge or energy flow between one-dimensional (1D) materials and nanotubes. This assists accurate prediction of their physical and chemical properties with respect to structural changes on the scale of encapsulated single atom thick changes. Recently developed separation techniques such as gel column chromatography and density gradient centrifugation offer opportunities for the efficient extraction of SWCNTs with particular chiralities and diameters from less refined mixtures. Here we demonstrate that chirality-refined SWCNTs filled with different categories of atomically thin HgTe NWs can be successfully separated based on a gel column chromatography method. The NWs display various types of HgTe arrangement, such as one atom-thick non-disordered linear chains, disordered atomic zig-zag chains, two-by-one atom-thick chains and bulk-like face-centered cubic structure encapsulated crystals. The properties of these composite materials were investigated by various optical and X-ray spectroscopic characterizations, including steady-state and time-resolved UV-Vis-infrared (terahertz (THz)) absorption, photoluminescence, Raman and X-ray photoelectron spectroscopy (XPS). A discrete change in the features of excitons and free charge carriers (i.e. plasmonic resonance) in SWCNTs as a result of HgTe NW filling was discovered. In addition, a strong dependence of such change on the SWCNT chiral structure and family type had also been demonstrated. In conclusion, we provided experimental evidence that atomic NW filling can modify the optical properties of SWCNTs. These findings lead to the possibility of more quantitative control on the performance of 1D nanomaterials and may open doors for new chemical applications. Ref: ACS Nano 2022, 16, 4, 6789–6800

Resume : In the field of nanotechnology, quantum dots (QDs) have been widely used in various applications and electronic devices in recent years, such as light-emitting diodes with quantum dots, quantum dot lasers, solar cells with quantum dots, single-electron transistors, single-photon sources and quantum computing. Due of the great importance of quantum dots as a class of quantum structures with excellent optical and electronic properties, several theoretical studies have been carried out to better understand the processes occurring in these structures. It is very important to control the mentioned properties with high accuracy. Ways to solve this range of problems include the change of the QD size, doping, and applying external fields. In this work we consider the quantum dot (QD) with the central and off-central acceptor impurities. We propose a model for determining the hole spectrum in the multiband model in the electric field with the presence of acceptor impurity. Also we determine interlevel absorption of electromagnetic wave in the quantum dot with acceptor impurity in the electric field. As a result of research, the holes spectrum in a multiband model in an electric field in different directions is determined. The dependence of the hole energy spectrum on the radius of the quantum dot and the location of the impurity from the center of the quantum dot is shown. The dependence of the energy spectrum of central and non-central acceptor impurity in the electric field and without it is obtained. Also in this work is shown that the absorption coefficient depends on the location of the impurity and the electrical application. In particular, the magnitude of the absorption coefficient in a quantum dot with and without impurity can be regulated by changing the energy field. It was researched that by changing the value of the electric field, one can tune the system to one or another wavelength.

Resume : Despite a few decades since their discovery, carbon nanostructures such as carbon nanotubes or graphene remain among the most intensively researched nanomaterials. They exhibit a considerable application potential spanning many areas of the R&D environment, ranging from microelectronics, via photonics, to biomedical engineering. Due to a strong structure-to-property relation, the characteristics of these materials can be tailored for specific applications. Alternatively, the properties of nanocarbon can be tuned by doping. This contribution will display how carbon nanotubes' electrical, thermoelectric, and optical properties can be greatly affected by exercising appropriate structural modification and doping. A thorough analysis of doping of carbon nanotubes using a broad spectrum of nitrogen-containing compounds will reveal the mechanics of the process. Furthermore, the presentation will show that not only the chemical composition of the dopant is essential, but its arrangement with respect to the doped material may also play a crucial role [1]. [1] B. Kumanek, K.Z. Milowska, L. Przypis, G. Stando, K. Matuszek, D. MacFarlane, M.C. Payne, D. Janas, Doping Engineering of Single-Walled Carbon Nanotubes by Nitrogen Compounds Using Basicity and Alignment, ACS Applied Materials & Interfaces (in press)

Resume : Higher manganese silicide (HMS) are prominent p-type thermoelectric materials for mid to high-temperature range power generation. In this work, we report the synthesis of highly oriented and partially substituted Mn(Si1-xAlx)1.74 single crystals which are melt-grown by the Bridgman method to attain a remarkable enhancement in zT~0.6 at 800K for the optimized single-crystal specimens directed perpendicular to c-axis. It was observed that Al as a p-type dopant exhibits a higher solubility in the single crystal which enhances the electrical conductivity and carrier concentration with a marginal decrease in the Seebeck coefficient to enhance the power factor significantly. Simultaneously, an enhanced point defect scattering resulted in a significant reduction of lattice thermal conductivity to result in a maximum zT~0.6 at 800K. Our results demonstrate the efficacy of aliovalent Al-doping in HMS single crystals for optimizing the carrier concentration and enhancing the phonon scattering, synergistically which can be also explored in other transition metal silicides for optimizing the thermoelectric transport properties.