Search

As money laundering is a global threat, with approximately US$800 billion to US$2 trillion being laundered yearly, it is important to come up with new and stronger solutions to combat illicit activities. Among the most exploited entities for money laundering are financial institutions, and more specifically banks. Therefore, the purpose of this thesis has been to investigate the usability of Artificial Intelligence (AI) as a decision-maker in the Anti-Money Laundering (AML) department to combat money laundering and terrorist financing. Moreover, the main research question was to explore the usability of AI as a decision-maker within AML operations. To enable this, an important aspect was to understand what regulatory and compliance requirements are demanded on AI from the AML departments. Thereafter, it was necessary to understand AI’s technical functionalities to facilitate the department’s daily operations. The literature review presented the fundamentals of AML and their precautionary actions. Furthermore, in the second phase of the literature review, the fundamentals of AI were investigated. Lastly, a combination of the topics of AML and AI were reviewed to identify techniques that the AML department can implement. Moreover, the methodology of the thesis consisted of a qualitative research design with an inductive approach. In the findings, a framework was constructed from the obtained results (second order themes). Subsequently, those themes were further developed into aggregated dimensions, which were extensively elaborated on in the conclusion. Thereafter, the dimensions were categorized into either the compliance criteria or the functional criteria. Also, the dimensions were presented in the order of priority based on how critical they are. To answer the main research question, AI is usable in the AML department as a decision-maker when considering the aggregated dimensions. Essentially, AI can successfully be implemented into the AML department’s daily oper, Då penningtvätt är ett globalt hot, där det uppskattas att det årligen tvättas 800 miljarder till 2 biljoner amerikanska dollar, är det viktigt att komma med nya och motståndskraftigare lösningar som bekämpar illegala verksamheter. Finansiella institutioner är bland de mest utsatta enheterna för penningtvätt, däribland banker som är mest exponerade. Därför var syftet med masterexamensarbetet att undersöka användbarheten av artificiell intelligens (AI) som beslutsfattare på avdelningen som bekämpar penningtvätt och finansiering av terrorism. Vidare var huvudfrågeställningen att utforska användbarheten av AI som beslutsfattare inom arbetsuppgifterna för avdelningen som bekämpar penningtvätt. För att möjliggöra detta var en viktig aspekt att förstå vilka lagstadgade- och efterlevnadskrav som ställdes på AI från avdelningen som bekämpar penningtvätt. Därefter måste man förstå AI:s tekniska funktioner för att underlätta avdelningens dagliga verksamhet. I litteraturstudien presenterades de grundläggande faktorerna för bekämpning av penningtvätt och deras försiktighetsåtgärder. Vidare i den andra fasen av litteraturstudien undersöktes de grundläggande faktorerna för AI. Slutligen granskades litteratur som behandlade en kombination av ämnena bekämpning av penningtvätt och AI för att identifiera tekniker som avdelningen för bekämpning av penningtvätt kan implementera. Dessutom bestod masterexamensarbetet av en kvalitativ forskningsdesign med en induktiv forskningsprocess. I resultatdelen framställdes ett ramverk utifrån de erhållna resultaten (andra ordningens teman). Därefter utvecklades dessa teman vidare till aggregerade dimensioner, som beskrevs utförligt i slutsatsen. Därpå blev dimensionerna kategoriserade i antingen efterlevnadskriterier eller funktionella kriterier. Dimensionerna presenterades även i prioritetsordning baserat på deras kritiska innebörd. Studien föreslog att AI är användbart på avdelningen som bekämpar penningtvätt som beslutsfattare när de aggregerad

We have investigated the optical properties of heterostructured InGaN platelets aiming at red emission, intended for use as nano-scaled light-emitting diodes. The focus is on the presence of non-radiative emission in the form of dark line defects. We have performed the study using hyperspectral cathodoluminescence imaging. The platelets were grown on a template consisting of InGaN pyramids, flattened by chemical mechanical polishing. These templates are defect free, whereas the dark line defects are introduced in the lower barrier and tend to propagate through all the subsequent layers, as revealed by the imaging of different layers in the structure. We conclude that the dark line defects are caused by stacking mismatch boundaries introduced by multiple seeding and step bunching at the edges of the as-polished, dome shaped templates. To avoid these defects, we suggest that the starting material must be flat rather than dome shaped., Funding Agencies|Lund Nano Lab [2022-02832]; Swedish Research Council [EM16-0024]; Foundation for Strategic Research; Knut and Alice Wallenberg foundation (KAW); Swedish Energy Agency; Crafoord foundation; NanoLund; Swedish Research Council (VR) [2021-00171, RIF21-0026]; Swedish Foundation for Strategic Research (SSF)

3D joint estimation can be used to track bodies in areas such as entertainment, sports, biomedicine and surveillance to identify bodies from video streams and images. This is most commonly done with multi-view solutions but researchers are currently spending a large amount of resources into developing mono-view solutions. The idea is to utilise neural networks to identify 3D joints by exploiting patterns and restrictions found in the human pose. Currently these systems are showing great results in controlled settings with good accuracy. However, for this to become a widespread technique it will be crucial for the systems to be able to perform with high accuracy in all types of settings. This thesis will focus on evaluating if current systems could be used to perform 3D estimations with high accuracy on movement analysis in sports settings. Based on a prestudy performing meaningful analysis in the area would require the system to perform with an accuracy of 4 cm. In order to evaluate the accuracy in this setting this thesis consists of three steps. Firstly two methods are picked by performing a prestudy of currently available monocular 3D joint estimation solutions. The accuracy of these two methods is then evaluated on two datasets, one which both have been trained on and another sports focused dataset which neither have been trained on. The sports dataset consisted of video sequences of movements from tennis, volleyball, basketball, badminton, football and rugby. In the last step a smoothing filter is applied on the results from the method that performed best on the sports dataset. This was done in order to further improve the accuracy of the system and evaluate the idea of using such techniques. The comparisons were made by measuring the mean per joint positional error for each of the images in the datasets. Both of the methods used were unable to reach a 4 cm accuracy on the sports dataset and thus are not suitable for this type of analysis in their current state, 3D estimering av leder i kroppen är en teknik som kan användas för att spåra och identifiera kroppar utifrån bilder och videoströmmar inom områden som underhållning, sport, biomedicin och övervakning. Detta görs oftast med flera kameror men forskare lägger för närvarande mycket resurser på att utveckla lösningar som bara använder en kamera, så kallade monocular system. Tanken är att använda neurala nätverk för att identifiera 3D leder genom att utnyttja mönster och restriktioner som finns i den mänskliga kroppen. För närvarande visar dessa system mycket lovande resultat i kontrollerade miljöer med mycket god noggrannhet. Men för att detta ska bli en utbredd teknik inom fler områden kommer det att vara avgörande för systemen att kunna prestera med hög noggrannhet i en större mängd miljöer med fler variationer. Denna avhandling fokuserar på att utvärdera om nuvarande metoder skulle kunna användas för att utföra 3D beräkningar av leder med tillräckligt hög noggrannhet för att kunna användas för rörelseanalys i sportmiljöer. Baserat på en förstudie inom det biomedicinska området skulle detta kräva att systemen kan arbeta med en noggrannhet på 4 cm. För att utvärdera noggrannheten i denna miljö, har metoden i denna avhandling delats upp i tre steg. Först kommer två metoder för 3D-uppskattning att väljas genom att utföra en förstudie av för närvarande tillgängliga monocular lösningar. Noggrannheten av dessa två utvärderas sedan på två dataset, ett som båda har tränats på och ett ytterligare dataset med fokus på sport som ingendera har tränats på. Sportdatasetet bestod av videosekvenser av rörelser från tennis, volleyboll, basket, badminton, fotboll och rugby. Detta är avgörande för det sista steget i metoden då ett filter appliceras på resultaten från den metod som presterade bäst på sportdatasetet. Detta gjordes för att ytterligare förbättra systemets noggrannhet och utvärdera idén med att använda sådana tekniker. Jämförelserna gjordes genom att mäta medelvärdet per ledp

Sammanfattning Dimensionering av traverskranbanor enligt Eurokod består av åtskilliga dimensionerings-villkor som bör uppfyllas inom brottgränstillstånd, bruksgränstillstånd och utmattning. Denna process är både tidskrävande och omständlig då det är många beräkningar gällande dimen-sioneringsvillkoren som behöver göras. I det här examensarbetet tas hjälpmedel fram för kranbanedimensionering i form av beräkningskalkylblad i Excel. Utöver detta undersöks det huruvida kritiskt vippningsmoment från FEM-Design kan vara ett alternativ till LTBeam, som är det rekommenderade programmet vid dimensionering av traverskranbanor. Detta resulterade i kalkylblad som fungerar att dimensionera traverskranbanor inom ramarna för det som de har avgränsats för. Resultaten gällande kritiska vippningsmomenten visade att FEM-Design gav ett 17,4 % lägre kritiskt vippningsmoment än LTBeam för det fall som studerats. Det högre kritiska vippningsmomentet från FEM-Design gav dock en ganska liten skillnad gällande utnyttjandegraden för vippningskontrollen jämfört med det kritiska vipp-ningsmomentet från LTBeam. Utnyttjandegraden blev 2,07 % högre med kritiskt vippnings-moment från FEM-Design än med kritiskt vippningsmoment från LTBeam. Med detta har slutsatsen dragits att FEM-Design är ett alternativ till LTBeam gällande beräkning av det kritiska vippningsmomentet till vippningskontrollen vid dimensionering av traverskranbanor., The design of overhead crane runways according to Eurocode outlines the design conditions that should be fulfilled within the ultimate limit state, serviceability limit state and during fatigue. This process is both time-consuming and cumbersome since there are many calculations regarding the design conditions that need to be made. In this thesis, aids are developed for the design of crane beams in the form of calculation spreadsheets in Excel. Additionally, it is investigated whether critical bending moment from FEM-Design can be an alternative to LTBeam, which is the recommended program for the design of crane beams. This resulted in spreadsheets that work to design crane beams within the limits of what they have been delineated for. The results indicate that FEM-Design gave a 17.4 % lower critical bending moment than LTBeam for the case studied. However, the higher critical bending moment from FEM-Design gave only a small difference regarding the degree of utilization of the control of lateral torsional buckling compared to LTBeam. The utilization rate was 2.07 % higher with the critical bending moment from FEM-Design than with LTBeam. According to this, the conclusion has been made that FEM-Design is an alternative to LTBeam for determining the critical bending moment for the control of lateral torsional buckling in the design of crane beams.

Compositionally graded channel AlGaN/GaN high electron mobility transistors (HEMTs) offer a promising route to improve device linearity, which is necessary for low-noise radio-frequency amplifiers. In this work, we demonstrate different grading profiles of a 10-nm-thick AlxGa1-xN channel from x = 0 to x = 0.1 using hot-wall metal-organic chemical vapor deposition (MOCVD). The growth process is developed by optimizing the channel grading and the channel-to-barrier transition. For this purpose, the Al-profiles and the interface sharpness, as determined from scanning transmission electron microscopy combined with energy-dispersive x-ray spectroscopy, are correlated with specific MOCVD process parameters. The results are linked to the channel properties (electron density, electron mobility, and sheet resistance) obtained by contactless Hall and terahertz optical Hall effect measurements coupled with simulations from solving self-consistently Poisson and Schrodinger equations. The impact of incorporating a thin AlN interlayer between the graded channel and the barrier layer on the HEMT properties is investigated and discussed. The optimized graded channel HEMT structure is found to have similarly high electron density (similar to 9 x 10(12) cm(-2)) as the non-graded conventional structure, though the mobility drops from similar to 2360 cm(2)/V s in the conventional to similar to 960 cm(2)/V s in the graded structure. The transconductance g(m) of the linearly graded channel HEMTs is shown to be flatter with smaller g(m) and g(m) as compared to the conventional non-graded channel HEMT implying improved device linearity. (c) 2023 Author(s). All article content, except where otherwise noted, is licensed under a Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/)., Funding Agencies|Swedish Governmental Agency for Innovation Systems (VINNOVA); Lund University; Linkoeping University; Chalmers University of Technology [2022-03139]; Ericsson; Epiluvac; FMV; Gotmic; Hexagem; Hitachi Energy; On Semiconductor; Region Skane SAAB; SweGaN; Volvo Cars; UMS; Swedish Research Council VR; Swedish Foundation for Strategic Research; Swedish Government Strategic Research Area in Materials Science on Functional Materials at Linkoeping University; KAW Foundation [2016-00889, 2022-04812]; Swedish Research Council and the Foundation [RIF14-055, EM16-0024, STP19-0008]; [2009-00971]; [2021-00171]; [RIF21-0026]

This study presents a novel approach to forming low-resistance ohmic contacts for AlGaN/GaN HEMTs. The optimized contacts exhibit an outstanding contact resistance of approximately 0.15 & omega;& BULL;mm. This is achieved by firstly recessing the barrier of the heterostructure to a depth beyond the channel. In this way, the channel region is exposed on the sidewall of the recess. The coverage of the Ti/Al/Ti ohmic metalization on the sidewall is ensured through tilting of the sample during evaporation. The annealing process is performed at a low temperature of 550 & DEG;C. The approach does not require precise control of the recess etching. Furthermore, the method is directly applicable to most barrier designs in terms of thickness and Al-concentration. The impact of recessed sidewall angle, thickness and ratio of Ti and Al layers, and the annealing procedure are investigated. Structural and chemical analyses of the interface between the ohmic contacts and epi-structure indicate the formation of ohmic contacts by the extraction of nitrogen from the epi-structure. The approach is demonstrated on HEMT-structures with two different barrier designs in terms of Al-concentration and barrier thickness. The study demonstrate large process window in regard to recess depth and duration of the annealing as well as high uniformity of the contact resistance across the samples, rendering the approach highly suitable for industrial production processes., Funding Agencies|Swedish Governmental Agency for Innovation Systems (VINNOVA); Linkoeping University; Chalmers University; ABB [2016-05190]; Epiluvac; Ericsson; FMV; Gotmic; Region Skane; Saab; SweGaN; Volvo Cars; United Monolithic Semiconductors (UMS); project Ultra-Compact AESA Technology for Autonomous Aircrafts; VINNOVA; Saab AB; project CoolHEMT; European Unions Horizon 2020 research and innovation programme [2017-04870]; Swedish Foundation for Strategic Research; Knut and Alice Wallenberg foundation; ARTEMI [823260]; Swedish National Infrastructure in Advanced Electron Microscopy - Swedish Research Council [EM16-0024, STP19-0008]; Foundation for strategic research; [2021-00171]; [RIF21-0026]

Structural defects are detrimental to the efficiency and quality of optoelectronic semiconductor devices. In this work, we study InGaN platelets with a quantum well structure intended for nano-LEDs emitting red light and how their optical properties, measured with cathodoluminescence, relate to the corresponding atomic structure. Through a method of spectroscopy-thinning-imaging, we demonstrate in plan-view how stacking mismatch boundaries intersect the quantum well in a pattern correlated with the observed diminished cathodoluminescence intensity. The results highlight the importance of avoiding stacking mismatch in small LED structures due to the relatively large region of non-radiative recombination caused by the mismatch boundaries., Funding Agencies|Swedish Governmental Agency for Innovation Systems (VINNOVA): Competence Center Program [2016-05190]; Swedish Government Strategic Research Area in Materials Science on Functional Materials at Linkoeping University, Faculty [2009-00971]; Swedish Research Council [EM16-0024]; Swedish Foundation for Strategic Research; Swedish National Infrastructure in Advanced Electron Microscopy [2022-02832, 2021-00171]; Swedish Foundation for Strategic Research [RIF21-0026]; Crafoord Foundation; NanoLund

Al xGa 1−xN/GaN high-electron-mobility transistor (HEMT) structures are key components in electronic devices operating at gigahertz or higher frequencies. In order to optimize such HEMT structures, understanding their electronic response at high frequencies and room temperature is required. Here, we present a study of the room temperature free charge carrier properties of the two-dimensional electron gas (2DEG) in HEMT structures with varying Al content in the Al xGa 1−xN barrier layers between x=0.07 and x=0.42⁠. We discuss and compare 2DEG sheet density, mobility, effective mass, sheet resistance, and scattering times, which are determined by theoretical calculations, contactless Hall effect, capacitance-voltage, Eddy current, and cavity-enhanced terahertz optical Hall effect (THz-OHE) measurements using a low-field permanent magnet (0.6 T). From our THz-OHE results, we observe that the measured mobility reduction from x=0.13 to x=0.42 is driven by the decrease in 2DEG scattering time, and not the change in effective mass. For x<0.42⁠, the 2DEG effective mass is found to be larger than for electrons in bulk GaN, which in turn, contributes to a decrease in the principally achievable mobility. From our theoretical calculations, we find that values close to 0.3 m0 can be explained by the combined effects of conduction band nonparabolicity, polarons, and hybridization of the electron wavefunction through penetration into the barrier layer., Funding Agencies|Foundation for Strategic Research, (2021-00171, RIF21-0026); Swedish Research Council VR, (2016-00889, 2022-04812); University of Nebraska Foundation; National Science Foundation, NSF, (NSF DMR 1808715); Air Force Office of Scientific Research, AFOSR, (FA9550-18-1-0360, FA9550-19-S-0003, FA9550-21-1-0259); Office of Experimental Program to Stimulate Competitive Research, EPSCoR, (OIA-2044049); J. A. Woollam Foundation; Stiftelsen för Strategisk Forskning, SSF, (2009-00971, EM16-0024, RIF14-055, STP19-0008); VINNOVA, (2022-03139); Knut och Alice Wallenbergs Stiftelse; Vetenskapsrådet, VR

Structural defects in Mg-doped GaN were analyzed using high-resolution scanning transmission electron microscopy combined with electron energy loss spectroscopy. The defects, in the shape of inverted pyramids, appear at high concentrations of incorporated Mg, which also lead to a reduction in free-hole concentration in Mg doped GaN. Detailed analysis pinpoints the arrangement of atoms in and around the defects and verify the presence of a well-defined layer of Mg at all facets, including the inclined facets. Our observations have resulted in a model of the pyramid-shaped defect, including structural displacements and compositional replacements, which is verified by image simulations. Finally, the total concentration of Mg atoms bound to these defects were evaluated, enabling a correlation between inactive and defect-bound dopants., Funding Agencies|Linkoping University

In traffic assignment, one of the key elements is the validation process. The validation allows the traffic planner to get an understanding of whether the model is working according to reality-based metrics, or if it must be improved. There are several different ways of collecting reality-based metrics, one these methods are called GPS probe data. Through this study, a survey as well as a case study has been performed to discover whether GPS probe data is a type of data suitable for this purpose. The study showed that using GPS probe data is a far more versatile method of collecting reality-based traffic data compared to conventional methods. This since information about a vehicle’s location is collected and stored for several points during a journey. These points allow analysis of vehicle movements for the entire network since the vehicles, acting like probes, gathers data during the entire trip. This enables analyses of entire trips and the ability to analyze the entire road network. However, since the GPS probe data is a rather raw type of data set, an extensive amount of data processing is required to carry out analyses of the data. Extensive and rather complicated data processing might reduce the number of traffic planners using the method, which put development of this relatively new method at risk. Therefore, there is a need of creating standardized and more user-friendly methods of processing GPS probe data if the benefits of the data type are going to be taken advantage of in the future of traffic planning., Att använda sig av så kallad GPS-baserad probedata i syftet av att validera trafikmodeller är en relativt ny och oprövad metod. Insamling av GPS-baserad probedata sker genom att fordon registrerar sin lokalisering vid flera punkter utmed sin resa. Metoden har potentialen att effektivisera och förbättra arbetet med att producera verklighetsförankrade och pålitliga trafikmodeller. Men för att nå dit krävs det omfattande arbete med att införa standardiserade metoder för den viktiga datahanteringen som krävs. Att producera trafikmodeller som representerar verkligheten är en vital del i arbetet med trafikplanering. Men för att veta huruvida en trafikmodell faktiskt liknar verkligheten behöver den jämföras med värden insamlade från verkliga mätningar, en process som kallas för validering. Att finna metoder för att effektivt kunna samla in data som representerar hela vägnätet är därmed av stor vikt. I denna studie jämfördes modelldata från en regional modell för Region Stockholm med GPS-baserad probedata för att testa probedatans tillämpbarhet som valideringskälla. Slutsatsen baserat på studiens resultat visar att det finns goda möjligheter att ta fram jämförbara data ur den GPS-baserade probedatan då mängden information som fås från varje resa är omfattande. Dock, på grund av informationens storlek krävs det omfattande databearbetning och utan standardiserade metoder finns det risk att arbetsprocessen blir ineffektiv och resultaten inkonsekventa. Vid validering av trafikmodeller jämförs parametrar såsom hastigheter, restider och ruttval mellan modell och den verklighetsbaserade datan. För insamling av data med syftet att validera trafikmodeller finns det många olika metoder. En metod som är relativt oprövad men har stor potential är så kallad GPS-baserad probedata. Detta är en insamlingsmetod som kan ge mätbara data för i princip hela vägnätet då det samlas in kontinuerligt via fordon i rörelse. En fördel med GPS-baserad probedata är därmed att en store del av vägnätet bl

The impact on the performance of GaN high electron mobility transistors (HEMTs) of in situ ammonia (NH3) pre-treatment prior to the deposition of silicon nitride (SiN) passivation with low-pressure chemical vapor deposition (LPCVD ) is investigated. Three different NH3 pre-treatment durations (0, 3, and 10 min) were compared in terms of interface properties and device performance. A reduction of oxygen (O) at the interface between SiN and epi-structure is detected by scanning transmission electron microscopy (STEM )-electron energy loss spectroscopy (EELS) measurements in the sample subjected to 10 min of pre-treatment. The samples subjected to NH3 pre-treatment show a reduced surface-related current dispersion of 9% (compared to 16% for the untreated sample), which is attributed to the reduction of O at the SiN/epi interface. Furthermore, NH3 pre-treatment for 10 min significantly improves the current dispersion uniformity from 14.5% to 1.9%. The reduced trapping effects result in a high output power of 3.4 W mm(-1) at 3 GHz (compared to 2.6 W mm(-1) for the untreated sample). These results demonstrate that the in situ NH3 pre-treatment before LPCVD of SiN passivation is critical and can effectively improves the large-signal microwave performance of GaN HEMTs., Funding Agencies|Swedish Governmental Agency for Innovation Systems (VINNOVA)Vinnova [2016-05190]; Linkoping University; Chalmers University; ABB; Epiluvac; EricssonEricsson; FMV; Gotmic; ON Semiconductor; Saab AB; SweGaN; United Monolithic Semiconductors (UMS); VINNOVAVinnova [2017-04870]; Swedish Research Council VRSwedish Research Council [2016-00889]; Swedish Foundation for Strategic Research (SSF)Swedish Foundation for Strategic Research [EM16-0024, STP19-0008]; European Unions Horizon 2020 research and innovation program [823260]; Knut and Alice Wallenberg foundationKnut & Alice Wallenberg Foundation

We report on the synthesis of vertical InP nanowire arrays on (001) InP and Si substrates using template-assisted vapour-liquid-solid growth. A thick silicon oxide layer was first deposited on the substrates. The samples were then patterned by electron beam lithography and deep dry etching through the oxide layer down to the substrate surface. Gold seed particles were subsequently deposited in the holes of the pattern by the use of pulse electrodeposition. The subsequent growth of nanowires by the vapour-liquid-solid method was guided towards the [001] direction by the patterned oxide template, and displayed a high growth yield with respect to the array of holes in the template. In order to confirm the versatility and robustness of the process, we have also demonstrated guided growth of InP nanowire p-n junctions and InP/InAs/InP nanowire heterostructures on (001) InP substrates. Our results show a promising route to monolithically integrate III-V nanowire heterostructure devices with commercially viable (001) silicon platforms.

As semiconducting nanowires have promising applications for materials science, nanoelectronics and photovoltaic devices, absolute control of their structure, morphology and composition is a prerequisite for their functionality. Nanowires, grown through the vapor-liquid-solid (VLS) method, exhibit a complex relationship between their liquid seed particle and what properties, such as the aforementioned ones, the wire adopts in the end. This means that compositional analysis, related to data on crystal structure and morphology, is crucial for better understanding. Transmission electron microscopy (TEM) plays an important role in atomically resolved studies of materials, including their elemental composition, which is of great use for evaluating growth methods and parameters for the nanowires. However, the conventional use of single projection TEM of the grown wires can neither reveal the 3D morphology nor potential rate of which the wire grows nor transient compositions during growth.In this work, we report on TEM studies of semiconducting nanowires. These studies are split into two parts: three-dimensional morphology through electron tomography, and detailed studies of the growth process using in-situ TEM. Both methods have been successful in providing additional information about the nanowires, otherwise impossible to obtain from conventional TEM. The tomographic studies have revealed the preference for forming Ga particles on GaAs nanowires’ {111}B facets and that the preferred facets formed in wurtzite core-shell nanowires differ between InAs and AlSb ({1-100} and {11-20} respectively). The surfaces of the reconstructed tomograms were illustrated through the novel technique of azimuthal mapping. In-situ TEM revealed rate-limiting steps of forming consecutive layers as well as it supplied compositional analysis of the seed particle during growth, which is of interest when producing nanowires with narrow specifications for an application. Also, a novel method of filt